Welcome to matlok
matlok
AI & ML interests
1000s of automated AI interactions a day (self-hosted 200+ GB VRAM and 300+ TFLOPS). Special thanks to all the amazing open source work/contributors that make this tech possible!
Recent Activity
liked a model about 3 hours ago
cyankiwi/Qwen3.6-35B-A3B-AWQ-4bit liked a model about 3 hours ago
cyankiwi/Qwen3.6-27B-AWQ-INT4 liked a dataset about 3 hours ago
open-agent-tools/open-toolsOrganizations
None yet
https://github.com/clonebench/BigCloneBench and https://github.com/jeffsvajlenko/BigCloneEval
https://zenodo.org/records/4247595
https://pytorch.org/blog/flexattention/
Note: BPE does not handle incremental patching like BLT
Patch start detected by entropy crossing a threshold
-
Qwen/QwQ-32B-Preview
Text Generation • 33B • Updated • 10.3k • • 1.74k -
bartowski/Qwen2.5-Coder-14B-Instruct-GGUF
Text Generation • 15B • Updated • 61.3k • 47 -
bartowski/Qwen2.5-Coder-32B-Instruct-GGUF
Text Generation • 33B • Updated • 36.5k • 106 -
bartowski/Qwen2.5-72B-Instruct-GGUF
Text Generation • 73B • Updated • 20.2k • 45
-
Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models
Paper • 2411.14257 • Published • 14 -
Scaling and evaluating sparse autoencoders
Paper • 2406.04093 • Published • 4 -
Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2
Paper • 2408.05147 • Published • 41 -
Disentangling Dense Embeddings with Sparse Autoencoders
Paper • 2408.00657 • Published • 1
-
Movie Gen: A Cast of Media Foundation Models
Paper • 2410.13720 • Published • 100 -
F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching
Paper • 2410.06885 • Published • 47 -
Flow Matching for Generative Modeling
Paper • 2210.02747 • Published • 4 -
Matcha-TTS: A fast TTS architecture with conditional flow matching
Paper • 2309.03199 • Published • 15
Python Alpaca instructions from leading AI research and tools repositories - focus is on "Manager level" understanding atm
-
matlok/python-text-copilot-training-instruct-ai-research-2024-02-11
Viewer • Updated • 130 • 1.79k -
matlok/python-text-copilot-training-instruct-ai-research-2024-02-10
Viewer • Updated • 123 • 4.21k -
matlok/python-text-copilot-training-instruct-ai-research-2024-02-03
Viewer • Updated • 2.67k • 646 • 1 -
matlok/python-text-copilot-training-instruct-ai-research-2024-01-27
Viewer • Updated • 43.1k • 159
Python code from leading AI research and tools repositories
there's many more on arxiv if you search for CLAP
-
Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation
Paper • 2211.06687 • Published • 6 -
EnCLAP: Combining Neural Audio Codec and Audio-Text Joint Embedding for Automated Audio Captioning
Paper • 2401.17690 • Published • 5 -
Amphion: An Open-Source Audio, Music and Speech Generation Toolkit
Paper • 2312.09911 • Published • 55 -
Audiobox: Unified Audio Generation with Natural Language Prompts
Paper • 2312.15821 • Published • 16
-
From Audio to Photoreal Embodiment: Synthesizing Humans in Conversations
Paper • 2401.01885 • Published • 28 -
Media2Face: Co-speech Facial Animation Generation With Multi-Modality Guidance
Paper • 2401.15687 • Published • 24 -
Unified-IO 2: Scaling Autoregressive Multimodal Models with Vision, Language, Audio, and Action
Paper • 2312.17172 • Published • 30 -
MouSi: Poly-Visual-Expert Vision-Language Models
Paper • 2401.17221 • Published • 9
There's usually interesting papers in the model cards on the leaderboard: https://huggingface.co/spaces/bigcode/bigcode-models-leaderboard
-
StarCoder: may the source be with you!
Paper • 2305.06161 • Published • 33 -
WizardCoder: Empowering Code Large Language Models with Evol-Instruct
Paper • 2306.08568 • Published • 33 -
SantaCoder: don't reach for the stars!
Paper • 2301.03988 • Published • 7 -
DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence
Paper • 2401.14196 • Published • 72
-
Improving Text Embeddings with Large Language Models
Paper • 2401.00368 • Published • 83 -
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Paper • 1810.04805 • Published • 28 -
Metadata Might Make Language Models Better
Paper • 2211.10086 • Published • 4 -
DecoderLens: Layerwise Interpretation of Encoder-Decoder Transformers
Paper • 2310.03686 • Published • 3
Large Multimodal Models
-
BLOOM: A 176B-Parameter Open-Access Multilingual Language Model
Paper • 2211.05100 • Published • 37 -
Contrastive Language-Image Pre-training for the Italian Language
Paper • 2108.08688 • Published • 2 -
IT5: Large-scale Text-to-text Pretraining for Italian Language Understanding and Generation
Paper • 2203.03759 • Published • 5 -
Spanish Pre-trained BERT Model and Evaluation Data
Paper • 2308.02976 • Published • 3
-
metavoiceio/metavoice-1B-v0.1
Text-to-Speech • Updated • 39 • 790 -
BASE TTS: Lessons from building a billion-parameter Text-to-Speech model on 100K hours of data
Paper • 2402.08093 • Published • 61 -
EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions
Paper • 2402.17485 • Published • 194 -
SWivid/F5-TTS
Text-to-Speech • Updated • 482k • 1.17k
-
In Search of Needles in a 10M Haystack: Recurrent Memory Finds What LLMs Miss
Paper • 2402.10790 • Published • 42 -
LongAgent: Scaling Language Models to 128k Context through Multi-Agent Collaboration
Paper • 2402.11550 • Published • 19 -
A Neural Conversational Model
Paper • 1506.05869 • Published • 2 -
Data Engineering for Scaling Language Models to 128K Context
Paper • 2402.10171 • Published • 25
-
AnyGPT: Unified Multimodal LLM with Discrete Sequence Modeling
Paper • 2402.12226 • Published • 45 -
M2-CLIP: A Multimodal, Multi-task Adapting Framework for Video Action Recognition
Paper • 2401.11649 • Published • 3 -
Gen4Gen: Generative Data Pipeline for Generative Multi-Concept Composition
Paper • 2402.15504 • Published • 21 -
EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions
Paper • 2402.17485 • Published • 194
-
A Generalized Language Model as the Combination of Skipped n-grams and Modified Kneser-Ney Smoothing
Paper • 1404.3377 • Published • 2 -
Skip-gram Language Modeling Using Sparse Non-negative Matrix Probability Estimation
Paper • 1412.1454 • Published • 2 -
Neural Text Generation from Structured Data with Application to the Biography Domain
Paper • 1603.07771 • Published • 2 -
Distributed Representations of Words and Phrases and their Compositionality
Paper • 1310.4546 • Published • 3
-
VideoPrism: A Foundational Visual Encoder for Video Understanding
Paper • 2402.13217 • Published • 40 -
EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions
Paper • 2402.17485 • Published • 194 -
Qwen/Qwen-VL-Chat
Text Generation • Updated • 89.3k • 383 -
MovieLLM: Enhancing Long Video Understanding with AI-Generated Movies
Paper • 2403.01422 • Published • 30
-
Improving Robustness for Joint Optimization of Camera Poses and Decomposed Low-Rank Tensorial Radiance Fields
Paper • 2402.13252 • Published • 19 -
RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputs
Paper • 2112.00724 • Published • 2 -
Enhancing NeRF akin to Enhancing LLMs: Generalizable NeRF Transformer with Mixture-of-View-Experts
Paper • 2308.11793 • Published • 2
-
Training Compute-Optimal Large Language Models
Paper • 2203.15556 • Published • 11 -
Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
Paper • 1909.08053 • Published • 5 -
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Paper • 1910.10683 • Published • 18 -
Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling
Paper • 2304.01373 • Published • 9
-
Lossless Acceleration for Seq2seq Generation with Aggressive Decoding
Paper • 2205.10350 • Published • 2 -
Blockwise Parallel Decoding for Deep Autoregressive Models
Paper • 1811.03115 • Published • 2 -
Fast Transformer Decoding: One Write-Head is All You Need
Paper • 1911.02150 • Published • 9 -
Sequence-Level Knowledge Distillation
Paper • 1606.07947 • Published • 2
-
Geometric Algebra Transformers
Paper • 2305.18415 • Published • 3 -
World Model on Million-Length Video And Language With RingAttention
Paper • 2402.08268 • Published • 40 -
Deep Unsupervised Learning using Nonequilibrium Thermodynamics
Paper • 1503.03585 • Published • 6 -
IDKiro/sdxs-512-0.9
Text-to-Image • Updated • 807 • 110
-
Resonance RoPE: Improving Context Length Generalization of Large Language Models
Paper • 2403.00071 • Published • 24 -
Scaling Laws of RoPE-based Extrapolation
Paper • 2310.05209 • Published • 8 -
Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models
Paper • 2404.12387 • Published • 40 -
OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework
Paper • 2404.14619 • Published • 126
-
Scaling Laws for Neural Language Models
Paper • 2001.08361 • Published • 10 -
An Empirical Model of Large-Batch Training
Paper • 1812.06162 • Published • 3 -
Measuring the Effects of Data Parallelism on Neural Network Training
Paper • 1811.03600 • Published • 2 -
Adafactor: Adaptive Learning Rates with Sublinear Memory Cost
Paper • 1804.04235 • Published • 2
-
Non-asymptotic oracle inequalities for the Lasso in high-dimensional mixture of experts
Paper • 2009.10622 • Published • 1 -
MoE-LLaVA: Mixture of Experts for Large Vision-Language Models
Paper • 2401.15947 • Published • 53 -
MoE-Mamba: Efficient Selective State Space Models with Mixture of Experts
Paper • 2401.04081 • Published • 74 -
MoE-Infinity: Activation-Aware Expert Offloading for Efficient MoE Serving
Paper • 2401.14361 • Published • 2
-
EVE: Efficient Vision-Language Pre-training with Masked Prediction and Modality-Aware MoE
Paper • 2308.11971 • Published • 2 -
HyperFormer: Enhancing Entity and Relation Interaction for Hyper-Relational Knowledge Graph Completion
Paper • 2308.06512 • Published • 2 -
Unraveling Complex Data Diversity in Underwater Acoustic Target Recognition through Convolution-based Mixture of Experts
Paper • 2402.11919 • Published • 2
-
Multi-view Contrastive Learning for Entity Typing over Knowledge Graphs
Paper • 2310.12008 • Published • 2 -
HyperFormer: Enhancing Entity and Relation Interaction for Hyper-Relational Knowledge Graph Completion
Paper • 2308.06512 • Published • 2 -
ARIEL: Adversarial Graph Contrastive Learning
Paper • 2208.06956 • Published • 2 -
RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval
Paper • 2401.18059 • Published • 48
-
Robust Mixture-of-Expert Training for Convolutional Neural Networks
Paper • 2308.10110 • Published • 2 -
HyperFormer: Enhancing Entity and Relation Interaction for Hyper-Relational Knowledge Graph Completion
Paper • 2308.06512 • Published • 2 -
Mobile V-MoEs: Scaling Down Vision Transformers via Sparse Mixture-of-Experts
Paper • 2309.04354 • Published • 16 -
Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints
Paper • 2212.05055 • Published • 6
-
Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
Paper • 1701.06538 • Published • 7 -
ST-MoE: Designing Stable and Transferable Sparse Expert Models
Paper • 2202.08906 • Published • 3 -
Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM
Paper • 2403.07816 • Published • 45
-
DocLLM: A layout-aware generative language model for multimodal document understanding
Paper • 2401.00908 • Published • 191 -
Unifying Vision, Text, and Layout for Universal Document Processing
Paper • 2212.02623 • Published • 12 -
Grounded Language-Image Pre-training
Paper • 2112.03857 • Published • 3 -
ConsistencyDet: Robust Object Detector with Denoising Paradigm of Consistency Model
Paper • 2404.07773 • Published • 1
-
Adding NVMe SSDs to Enable and Accelerate 100B Model Fine-tuning on a Single GPU
Paper • 2403.06504 • Published • 56 -
Token-Level Adaptation of LoRA Adapters for Downstream Task Generalization
Paper • 2311.10847 • Published • 2 -
PERL: Parameter Efficient Reinforcement Learning from Human Feedback
Paper • 2403.10704 • Published • 60
-
Mixture-of-Supernets: Improving Weight-Sharing Supernet Training with Architecture-Routed Mixture-of-Experts
Paper • 2306.04845 • Published • 4 -
Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networks
Paper • 2306.04073 • Published • 2 -
Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM
Paper • 2403.07816 • Published • 45 -
Unified Scaling Laws for Routed Language Models
Paper • 2202.01169 • Published • 2
-
Vulnerability Analysis of Transformer-based Optical Character Recognition to Adversarial Attacks
Paper • 2311.17128 • Published • 2 -
Data Generation for Post-OCR correction of Cyrillic handwriting
Paper • 2311.15896 • Published • 4 -
An End-to-End OCR Framework for Robust Arabic-Handwriting Recognition using a Novel Transformers-based Model and an Innovative 270 Million-Words Multi-Font Corpus of Classical Arabic with Diacritics
Paper • 2208.11484 • Published • 3 -
Transformer based Urdu Handwritten Text Optical Character Reader
Paper • 2206.04575 • Published • 2
-
Character Queries: A Transformer-based Approach to On-Line Handwritten Character Segmentation
Paper • 2309.03072 • Published • 2 -
Prompt me a Dataset: An investigation of text-image prompting for historical image dataset creation using foundation models
Paper • 2309.01674 • Published • 2 -
Segment Anything
Paper • 2304.02643 • Published • 6
-
Disentangling Writer and Character Styles for Handwriting Generation
Paper • 2303.14736 • Published • 3 -
A Transformer Architecture for Online Gesture Recognition of Mathematical Expressions
Paper • 2211.02643 • Published • 2 -
A tailored Handwritten-Text-Recognition System for Medieval Latin
Paper • 2308.09368 • Published • 3 -
Scalable handwritten text recognition system for lexicographic sources of under-resourced languages and alphabets
Paper • 2303.16256 • Published • 2
GitHub: https://github.com/Planet-AI-GmbH/tfaip-hybrid-ctc-s2s and math: https://storage.googleapis.com/mathwriting_data/mathwriting-2024.tgz
-
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Paper • 1810.04805 • Published • 28 -
Transformers Can Achieve Length Generalization But Not Robustly
Paper • 2402.09371 • Published • 14 -
Triple-Encoders: Representations That Fire Together, Wire Together
Paper • 2402.12332 • Published • 2 -
BERTs are Generative In-Context Learners
Paper • 2406.04823 • Published • 1
-
Pretraining-Based Natural Language Generation for Text Summarization
Paper • 1902.09243 • Published • 2 -
Unified Vision-Language Pre-Training for Image Captioning and VQA
Paper • 1909.11059 • Published • 2 -
When LLMs are Unfit Use FastFit: Fast and Effective Text Classification with Many Classes
Paper • 2404.12365 • Published • 2
-
Representing Online Handwriting for Recognition in Large Vision-Language Models
Paper • 2402.15307 • Published • 3 -
Evaluating Sequence-to-Sequence Models for Handwritten Text Recognition
Paper • 1903.07377 • Published • 2 -
Enhancing Document Information Analysis with Multi-Task Pre-training: A Robust Approach for Information Extraction in Visually-Rich Documents
Paper • 2310.16527 • Published • 2 -
Detecting and recognizing characters in Greek papyri with YOLOv8, DeiT and SimCLR
Paper • 2401.12513 • Published • 1
-
mT5: A massively multilingual pre-trained text-to-text transformer
Paper • 2010.11934 • Published • 4 -
mSLAM: Massively multilingual joint pre-training for speech and text
Paper • 2202.01374 • Published • 2 -
DeepNet: Scaling Transformers to 1,000 Layers
Paper • 2203.00555 • Published • 2 -
SONAR: Sentence-Level Multimodal and Language-Agnostic Representations
Paper • 2308.11466 • Published • 1
-
Enhancing Document Information Analysis with Multi-Task Pre-training: A Robust Approach for Information Extraction in Visually-Rich Documents
Paper • 2310.16527 • Published • 2 -
DocLLM: A layout-aware generative language model for multimodal document understanding
Paper • 2401.00908 • Published • 191 -
Unifying Vision, Text, and Layout for Universal Document Processing
Paper • 2212.02623 • Published • 12
-
DragAnything: Motion Control for Anything using Entity Representation
Paper • 2403.07420 • Published • 14 -
Capabilities of Gemini Models in Medicine
Paper • 2404.18416 • Published • 25 -
MMWorld: Towards Multi-discipline Multi-faceted World Model Evaluation in Videos
Paper • 2406.08407 • Published • 28
-
Synth^2: Boosting Visual-Language Models with Synthetic Captions and Image Embeddings
Paper • 2403.07750 • Published • 23 -
Gaussian Frosting: Editable Complex Radiance Fields with Real-Time Rendering
Paper • 2403.14554 • Published • 14 -
ObjectDrop: Bootstrapping Counterfactuals for Photorealistic Object Removal and Insertion
Paper • 2403.18818 • Published • 28
-
Wide Residual Networks
Paper • 1605.07146 • Published • 2 -
Characterizing signal propagation to close the performance gap in unnormalized ResNets
Paper • 2101.08692 • Published • 2 -
Pareto-Optimal Quantized ResNet Is Mostly 4-bit
Paper • 2105.03536 • Published • 3 -
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations
Paper • 2106.01548 • Published • 2
-
Image Segmentation using U-Net Architecture for Powder X-ray Diffraction Images
Paper • 2310.16186 • Published • 2 -
H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes
Paper • 1709.07330 • Published • 2 -
Deep LOGISMOS: Deep Learning Graph-based 3D Segmentation of Pancreatic Tumors on CT scans
Paper • 1801.08599 • Published • 2 -
RTSeg: Real-time Semantic Segmentation Comparative Study
Paper • 1803.02758 • Published • 2
-
MovieLLM: Enhancing Long Video Understanding with AI-Generated Movies
Paper • 2403.01422 • Published • 30 -
VisionGPT-3D: A Generalized Multimodal Agent for Enhanced 3D Vision Understanding
Paper • 2403.09530 • Published • 10 -
VidToMe: Video Token Merging for Zero-Shot Video Editing
Paper • 2312.10656 • Published • 11 -
TC4D: Trajectory-Conditioned Text-to-4D Generation
Paper • 2403.17920 • Published • 18
-
Generalizability vs. Robustness: Adversarial Examples for Medical Imaging
Paper • 1804.00504 • Published • 2 -
Evaluating Transformer-based Semantic Segmentation Networks for Pathological Image Segmentation
Paper • 2108.11993 • Published • 2 -
From Modern CNNs to Vision Transformers: Assessing the Performance, Robustness, and Classification Strategies of Deep Learning Models in Histopathology
Paper • 2204.05044 • Published • 2
-
3D Medical Image Segmentation based on multi-scale MPU-Net
Paper • 2307.05799 • Published • 2 -
Joint Liver and Hepatic Lesion Segmentation in MRI using a Hybrid CNN with Transformer Layers
Paper • 2201.10981 • Published • 2 -
Using Multi-scale SwinTransformer-HTC with Data augmentation in CoNIC Challenge
Paper • 2202.13588 • Published • 2
-
Attention Swin U-Net: Cross-Contextual Attention Mechanism for Skin Lesion Segmentation
Paper • 2210.16898 • Published • 2 -
Cross-Shaped Windows Transformer with Self-supervised Pretraining for Clinically Significant Prostate Cancer Detection in Bi-parametric MRI
Paper • 2305.00385 • Published • 2 -
Event Camera Demosaicing via Swin Transformer and Pixel-focus Loss
Paper • 2404.02731 • Published • 1
-
CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows
Paper • 2107.00652 • Published • 2 -
Glyph-ByT5: A Customized Text Encoder for Accurate Visual Text Rendering
Paper • 2403.09622 • Published • 17 -
Veagle: Advancements in Multimodal Representation Learning
Paper • 2403.08773 • Published • 10 -
mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality
Paper • 2304.14178 • Published • 3
-
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
Paper • 2103.14030 • Published • 5 -
A Novel Transformer Based Semantic Segmentation Scheme for Fine-Resolution Remote Sensing Images
Paper • 2104.12137 • Published • 2 -
Self-Supervised Learning with Swin Transformers
Paper • 2105.04553 • Published • 3 -
Evaluating Transformer-based Semantic Segmentation Networks for Pathological Image Segmentation
Paper • 2108.11993 • Published • 2
-
Self-Supervised Vision Transformers Learn Visual Concepts in Histopathology
Paper • 2203.00585 • Published • 3 -
Emerging Properties in Self-Supervised Vision Transformers
Paper • 2104.14294 • Published • 4 -
DreamScene360: Unconstrained Text-to-3D Scene Generation with Panoramic Gaussian Splatting
Paper • 2404.06903 • Published • 21 -
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32
-
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
Chain-of-Thought Reasoning Without Prompting
Paper • 2402.10200 • Published • 110 -
LLM2LLM: Boosting LLMs with Novel Iterative Data Enhancement
Paper • 2403.15042 • Published • 27 -
LIMA: Less Is More for Alignment
Paper • 2305.11206 • Published • 27
-
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
SELF: Language-Driven Self-Evolution for Large Language Model
Paper • 2310.00533 • Published • 2 -
QLoRA: Efficient Finetuning of Quantized LLMs
Paper • 2305.14314 • Published • 61 -
QA-LoRA: Quantization-Aware Low-Rank Adaptation of Large Language Models
Paper • 2309.14717 • Published • 46
-
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
SELF: Language-Driven Self-Evolution for Large Language Model
Paper • 2310.00533 • Published • 2 -
Bigger, Better, Faster: Human-level Atari with human-level efficiency
Paper • 2305.19452 • Published • 5 -
DeepSeek-Prover-V1.5: Harnessing Proof Assistant Feedback for Reinforcement Learning and Monte-Carlo Tree Search
Paper • 2408.08152 • Published • 62
-
SELF: Language-Driven Self-Evolution for Large Language Model
Paper • 2310.00533 • Published • 2 -
GrowLength: Accelerating LLMs Pretraining by Progressively Growing Training Length
Paper • 2310.00576 • Published • 2 -
A Pretrainer's Guide to Training Data: Measuring the Effects of Data Age, Domain Coverage, Quality, & Toxicity
Paper • 2305.13169 • Published • 4 -
Transformers Can Achieve Length Generalization But Not Robustly
Paper • 2402.09371 • Published • 14
-
JoMA: Demystifying Multilayer Transformers via JOint Dynamics of MLP and Attention
Paper • 2310.00535 • Published • 2 -
Solving Challenging Math Word Problems Using GPT-4 Code Interpreter with Code-based Self-Verification
Paper • 2308.07921 • Published • 24 -
AutoNumerics-Zero: Automated Discovery of State-of-the-Art Mathematical Functions
Paper • 2312.08472 • Published • 2 -
Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems
Paper • 2408.16293 • Published • 27
-
MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data
Paper • 2304.08247 • Published • 2 -
Structural Similarities Between Language Models and Neural Response Measurements
Paper • 2306.01930 • Published • 2 -
Multimodal ChatGPT for Medical Applications: an Experimental Study of GPT-4V
Paper • 2310.19061 • Published • 8 -
Question-Answering Model for Schizophrenia Symptoms and Their Impact on Daily Life using Mental Health Forums Data
Paper • 2310.00448 • Published
-
Demystifying CLIP Data
Paper • 2309.16671 • Published • 20 -
Model Stock: All we need is just a few fine-tuned models
Paper • 2403.19522 • Published • 15 -
Bigger is not Always Better: Scaling Properties of Latent Diffusion Models
Paper • 2404.01367 • Published • 22 -
On the Scalability of Diffusion-based Text-to-Image Generation
Paper • 2404.02883 • Published • 19
-
Unified Functional Hashing in Automatic Machine Learning
Paper • 2302.05433 • Published • 2 -
Self-Discover: Large Language Models Self-Compose Reasoning Structures
Paper • 2402.03620 • Published • 117 -
Semi-Supervised Semantic Segmentation using Redesigned Self-Training for White Blood Cells
Paper • 2401.07278 • Published • 2
-
AutoNumerics-Zero: Automated Discovery of State-of-the-Art Mathematical Functions
Paper • 2312.08472 • Published • 2 -
MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
Paper • 2403.14624 • Published • 53 -
ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline
Paper • 2404.02893 • Published • 21 -
Rho-1: Not All Tokens Are What You Need
Paper • 2404.07965 • Published • 94
-
Unlocking the conversion of Web Screenshots into HTML Code with the WebSight Dataset
Paper • 2403.09029 • Published • 57 -
HuggingFaceM4/WebSight
Viewer • Updated • 2.75M • 14k • 393 -
HuggingFaceM4/VLM_WebSight_finetuned
Text Generation • 8B • Updated • 202 • 192 -
laion/filtered-wit
Viewer • Updated • 2.8M • 6.27k • 11
-
Functional Interpolation for Relative Positions Improves Long Context Transformers
Paper • 2310.04418 • Published • 4 -
SPBERT: An Efficient Pre-training BERT on SPARQL Queries for Question Answering over Knowledge Graphs
Paper • 2106.09997 • Published • 2 -
Neural Machine Translation of Rare Words with Subword Units
Paper • 1508.07909 • Published • 4 -
A Multimodal Approach to Device-Directed Speech Detection with Large Language Models
Paper • 2403.14438 • Published • 2
-
Veagle: Advancements in Multimodal Representation Learning
Paper • 2403.08773 • Published • 10 -
mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality
Paper • 2304.14178 • Published • 3 -
Chart-based Reasoning: Transferring Capabilities from LLMs to VLMs
Paper • 2403.12596 • Published • 12 -
LLaVA-UHD: an LMM Perceiving Any Aspect Ratio and High-Resolution Images
Paper • 2403.11703 • Published • 17
-
Vision Transformer with Quadrangle Attention
Paper • 2303.15105 • Published • 2 -
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
Paper • 2103.14030 • Published • 5 -
MAFormer: A Transformer Network with Multi-scale Attention Fusion for Visual Recognition
Paper • 2209.01620 • Published • 2 -
CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows
Paper • 2107.00652 • Published • 2
-
Exploring Large Language Models' Cognitive Moral Development through Defining Issues Test
Paper • 2309.13356 • Published • 38 -
Unveiling Safety Vulnerabilities of Large Language Models
Paper • 2311.04124 • Published • 9 -
TrustLLM: Trustworthiness in Large Language Models
Paper • 2401.05561 • Published • 69 -
Evaluating Frontier Models for Dangerous Capabilities
Paper • 2403.13793 • Published • 7
-
Qwen Technical Report
Paper • 2309.16609 • Published • 38 -
Qwen-VL: A Frontier Large Vision-Language Model with Versatile Abilities
Paper • 2308.12966 • Published • 12 -
Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models
Paper • 2311.07919 • Published • 9 -
Audio Dialogues: Dialogues dataset for audio and music understanding
Paper • 2404.07616 • Published • 15
significant improvements in zero-shot performance require exponentially more data, following a log-linear scaling trend
-
MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries
Paper • 2401.15391 • Published • 6 -
RAFT: Adapting Language Model to Domain Specific RAG
Paper • 2403.10131 • Published • 72 -
Superposition Prompting: Improving and Accelerating Retrieval-Augmented Generation
Paper • 2404.06910 • Published • 3 -
Stylus: Automatic Adapter Selection for Diffusion Models
Paper • 2404.18928 • Published • 15
-
Streaming Transformer ASR with Blockwise Synchronous Beam Search
Paper • 2006.14941 • Published • 2 -
A Multimodal Approach to Device-Directed Speech Detection with Large Language Models
Paper • 2403.14438 • Published • 2 -
SONAR: Sentence-Level Multimodal and Language-Agnostic Representations
Paper • 2308.11466 • Published • 1
-
Neural Machine Translation of Rare Words with Subword Units
Paper • 1508.07909 • Published • 4 -
A Formal Perspective on Byte-Pair Encoding
Paper • 2306.16837 • Published • 3 -
Byte-Pair Encoding for Text-to-SQL Generation
Paper • 1910.08962 • Published • 2 -
Pattern Discovery in Time Series with Byte Pair Encoding
Paper • 2106.00614 • Published • 2
-
Recurrent Drafter for Fast Speculative Decoding in Large Language Models
Paper • 2403.09919 • Published • 21 -
SpecInfer: Accelerating Generative LLM Serving with Speculative Inference and Token Tree Verification
Paper • 2305.09781 • Published • 4 -
Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters
Paper • 2408.04093 • Published • 5
-
microsoft/table-transformer-detection
Object Detection • 28.8M • Updated • 3.08M • 421 -
RUCKBReasoning/TableLLM-13b
Text Generation • 13B • Updated • 108 • 32 -
RUCKBReasoning/TableLLM-7b
Text Generation • 7B • Updated • 57 • 16 -
TahaDouaji/detr-doc-table-detection
Object Detection • 41.6M • Updated • 234k • 63
-
GaussianFlow: Splatting Gaussian Dynamics for 4D Content Creation
Paper • 2403.12365 • Published • 12 -
RadSplat: Radiance Field-Informed Gaussian Splatting for Robust Real-Time Rendering with 900+ FPS
Paper • 2403.13806 • Published • 19 -
Gaussian Frosting: Editable Complex Radiance Fields with Real-Time Rendering
Paper • 2403.14554 • Published • 14 -
DreamPolisher: Towards High-Quality Text-to-3D Generation via Geometric Diffusion
Paper • 2403.17237 • Published • 11
-
End-to-End Object Detection with Transformers
Paper • 2005.12872 • Published • 7 -
COCONut: Modernizing COCO Segmentation
Paper • 2404.08639 • Published • 30 -
Grounded Language-Image Pre-training
Paper • 2112.03857 • Published • 3 -
Florence-2: Advancing a Unified Representation for a Variety of Vision Tasks
Paper • 2311.06242 • Published • 96
-
GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
Paper • 2305.13245 • Published • 6 -
Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models
Paper • 2404.12387 • Published • 40 -
OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework
Paper • 2404.14619 • Published • 126 -
Qwen2 Technical Report
Paper • 2407.10671 • Published • 172
-
MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
Paper • 2403.14624 • Published • 53 -
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?
Paper • 2407.01284 • Published • 81 -
DynaMath: A Dynamic Visual Benchmark for Evaluating Mathematical Reasoning Robustness of Vision Language Models
Paper • 2411.00836 • Published • 15
-
Explorative Inbetweening of Time and Space
Paper • 2403.14611 • Published • 13 -
StyleCineGAN: Landscape Cinemagraph Generation using a Pre-trained StyleGAN
Paper • 2403.14186 • Published • 10 -
Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction
Paper • 2403.18795 • Published • 20
-
Python Fuzzing for Trustworthy Machine Learning Frameworks
Paper • 2403.12723 • Published • 2 -
Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?
Paper • 2404.03411 • Published • 10 -
Teams of LLM Agents can Exploit Zero-Day Vulnerabilities
Paper • 2406.01637 • Published • 2 -
AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases
Paper • 2407.12784 • Published • 51
-
CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing
Paper • 2305.11738 • Published • 9 -
CriticBench: Benchmarking LLMs for Critique-Correct Reasoning
Paper • 2402.14809 • Published • 3 -
DRLC: Reinforcement Learning with Dense Rewards from LLM Critic
Paper • 2401.07382 • Published • 2
-
V-STaR: Training Verifiers for Self-Taught Reasoners
Paper • 2402.06457 • Published • 9 -
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
Language Models as Compilers: Simulating Pseudocode Execution Improves Algorithmic Reasoning in Language Models
Paper • 2404.02575 • Published • 50
-
ThemeStation: Generating Theme-Aware 3D Assets from Few Exemplars
Paper • 2403.15383 • Published • 15 -
FlexiDreamer: Single Image-to-3D Generation with FlexiCubes
Paper • 2404.00987 • Published • 23 -
MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs
Paper • 2402.15627 • Published • 36 -
Interactive3D: Create What You Want by Interactive 3D Generation
Paper • 2404.16510 • Published • 21
-
FollowIR: Evaluating and Teaching Information Retrieval Models to Follow Instructions
Paper • 2403.15246 • Published • 11 -
Noise-Aware Training of Layout-Aware Language Models
Paper • 2404.00488 • Published • 10 -
SnapKV: LLM Knows What You are Looking for Before Generation
Paper • 2404.14469 • Published • 27
-
Beyond A*: Better Planning with Transformers via Search Dynamics Bootstrapping
Paper • 2402.14083 • Published • 47 -
GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
Paper • 2305.13245 • Published • 6 -
Training a T5 Using Lab-sized Resources
Paper • 2208.12097 • Published • 1 -
Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints
Paper • 2212.05055 • Published • 6
-
Prompt me a Dataset: An investigation of text-image prompting for historical image dataset creation using foundation models
Paper • 2309.01674 • Published • 2 -
Segment Anything
Paper • 2304.02643 • Published • 6 -
EgoLifter: Open-world 3D Segmentation for Egocentric Perception
Paper • 2403.18118 • Published • 12 -
A Multimodal Automated Interpretability Agent
Paper • 2404.14394 • Published • 23
-
LIMA: Less Is More for Alignment
Paper • 2305.11206 • Published • 27 -
Garment3DGen: 3D Garment Stylization and Texture Generation
Paper • 2403.18816 • Published • 25 -
EgoLifter: Open-world 3D Segmentation for Egocentric Perception
Paper • 2403.18118 • Published • 12 -
The Unreasonable Ineffectiveness of the Deeper Layers
Paper • 2403.17887 • Published • 83
Could also use a dna repo like: https://github.com/koeng101/dnadesign
-
One-step Diffusion with Distribution Matching Distillation
Paper • 2311.18828 • Published • 3 -
The Unreasonable Ineffectiveness of the Deeper Layers
Paper • 2403.17887 • Published • 83 -
Condition-Aware Neural Network for Controlled Image Generation
Paper • 2404.01143 • Published • 13 -
Locating and Editing Factual Associations in GPT
Paper • 2202.05262 • Published • 1
-
The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
Paper • 1801.03924 • Published • 2 -
Fine-Tuning Language Models from Human Preferences
Paper • 1909.08593 • Published • 4 -
Training Verifiers to Solve Math Word Problems
Paper • 2110.14168 • Published • 7 -
Learning Transferable Visual Models From Natural Language Supervision
Paper • 2103.00020 • Published • 22
-
BioMedLM: A 2.7B Parameter Language Model Trained On Biomedical Text
Paper • 2403.18421 • Published • 23 -
Long-form factuality in large language models
Paper • 2403.18802 • Published • 26 -
stanford-crfm/BioMedLM
Text Generation • Updated • 5.78k • 454 -
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
Paper • 2305.18290 • Published • 66
-
Can large language models explore in-context?
Paper • 2403.15371 • Published • 33 -
Long-context LLMs Struggle with Long In-context Learning
Paper • 2404.02060 • Published • 37 -
PIQA: Reasoning about Physical Commonsense in Natural Language
Paper • 1911.11641 • Published • 5 -
AQuA: A Benchmarking Tool for Label Quality Assessment
Paper • 2306.09467 • Published • 1
-
TextCraftor: Your Text Encoder Can be Image Quality Controller
Paper • 2403.18978 • Published • 15 -
InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation
Paper • 2404.02733 • Published • 22 -
OmniFusion Technical Report
Paper • 2404.06212 • Published • 77 -
Transferable and Principled Efficiency for Open-Vocabulary Segmentation
Paper • 2404.07448 • Published • 12
-
LITA: Language Instructed Temporal-Localization Assistant
Paper • 2403.19046 • Published • 19 -
Snap-it, Tap-it, Splat-it: Tactile-Informed 3D Gaussian Splatting for Reconstructing Challenging Surfaces
Paper • 2403.20275 • Published • 10 -
Condition-Aware Neural Network for Controlled Image Generation
Paper • 2404.01143 • Published • 13 -
CantTalkAboutThis: Aligning Language Models to Stay on Topic in Dialogues
Paper • 2404.03820 • Published • 25
-
Learning Trajectory Preferences for Manipulators via Iterative Improvement
Paper • 1306.6294 • Published • 3 -
MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection
Paper • 2403.19888 • Published • 12 -
RL for Consistency Models: Faster Reward Guided Text-to-Image Generation
Paper • 2404.03673 • Published • 15 -
PhysDreamer: Physics-Based Interaction with 3D Objects via Video Generation
Paper • 2404.13026 • Published • 24
https://web.stanford.edu/class/archive/stats/stats200/stats200.1172/Lecture24.pdf
-
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
Paper • 2305.18290 • Published • 66 -
HyperCLOVA X Technical Report
Paper • 2404.01954 • Published • 25 -
Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization
Paper • 2404.09956 • Published • 11 -
Learn Your Reference Model for Real Good Alignment
Paper • 2404.09656 • Published • 90
-
The Curious Case of Neural Text Degeneration
Paper • 1904.09751 • Published • 3 -
Getting it Right: Improving Spatial Consistency in Text-to-Image Models
Paper • 2404.01197 • Published • 31 -
BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions
Paper • 1905.10044 • Published • 2 -
PIQA: Reasoning about Physical Commonsense in Natural Language
Paper • 1911.11641 • Published • 5
Spectrogram Prediction Network
As in Tacotron, mel spectrograms are computed through a short-
time Fourier transform (STFT) ... and a Hann window func
-
Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions
Paper • 1712.05884 • Published • 3 -
Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization
Paper • 2404.09956 • Published • 11 -
Music Consistency Models
Paper • 2404.13358 • Published • 14 -
FlashSpeech: Efficient Zero-Shot Speech Synthesis
Paper • 2404.14700 • Published • 32
-
WhisperX: Time-Accurate Speech Transcription of Long-Form Audio
Paper • 2303.00747 • Published • 6 -
Custom Data Augmentation for low resource ASR using Bark and Retrieval-Based Voice Conversion
Paper • 2311.14836 • Published • 2 -
SONAR: Sentence-Level Multimodal and Language-Agnostic Representations
Paper • 2308.11466 • Published • 1 -
W2v-BERT: Combining Contrastive Learning and Masked Language Modeling for Self-Supervised Speech Pre-Training
Paper • 2108.06209 • Published • 1
https://kyutai.org/
-
MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection
Paper • 2403.19888 • Published • 12 -
Measuring Style Similarity in Diffusion Models
Paper • 2404.01292 • Published • 16 -
GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding
Paper • 1804.07461 • Published • 4 -
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15
-
Gecko: Versatile Text Embeddings Distilled from Large Language Models
Paper • 2403.20327 • Published • 47 -
2D Matryoshka Sentence Embeddings
Paper • 2402.14776 • Published • 8 -
Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference
Paper • 2412.13663 • Published • 163
Gradient-based surface extraction method
-
Non-Autoregressive Neural Machine Translation
Paper • 1711.02281 • Published • 1 -
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation
Paper • 2107.07651 • Published • 1 -
OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
Paper • 2404.07972 • Published • 52 -
AgentOhana: Design Unified Data and Training Pipeline for Effective Agent Learning
Paper • 2402.15506 • Published • 18
-
WavLLM: Towards Robust and Adaptive Speech Large Language Model
Paper • 2404.00656 • Published • 11 -
CameraCtrl: Enabling Camera Control for Text-to-Video Generation
Paper • 2404.02101 • Published • 24 -
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1
-
Noise-Aware Training of Layout-Aware Language Models
Paper • 2404.00488 • Published • 10 -
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking
Paper • 2204.08387 • Published • 8 -
LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding
Paper • 2012.14740 • Published • 3 -
LayoutLM: Pre-training of Text and Layout for Document Image Understanding
Paper • 1912.13318 • Published • 5
-
Noise-Aware Training of Layout-Aware Language Models
Paper • 2404.00488 • Published • 10 -
FormNet: Structural Encoding beyond Sequential Modeling in Form Document Information Extraction
Paper • 2203.08411 • Published • 1 -
FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extraction
Paper • 2305.02549 • Published • 7 -
ETC: Encoding Long and Structured Inputs in Transformers
Paper • 2004.08483 • Published • 1
-
Streaming Dense Video Captioning
Paper • 2404.01297 • Published • 13 -
PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning
Paper • 2404.16994 • Published • 37 -
ShareGPT4Video: Improving Video Understanding and Generation with Better Captions
Paper • 2406.04325 • Published • 74
-
FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extraction
Paper • 2305.02549 • Published • 7 -
FormNet: Structural Encoding beyond Sequential Modeling in Form Document Information Extraction
Paper • 2203.08411 • Published • 1 -
ETC: Encoding Long and Structured Inputs in Transformers
Paper • 2004.08483 • Published • 1 -
LongT5: Efficient Text-To-Text Transformer for Long Sequences
Paper • 2112.07916 • Published • 2
-
Long-context LLMs Struggle with Long In-context Learning
Paper • 2404.02060 • Published • 37 -
Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks
Paper • 2211.12588 • Published • 3 -
StructLM: Towards Building Generalist Models for Structured Knowledge Grounding
Paper • 2402.16671 • Published • 27 -
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model
Paper • 2404.04167 • Published • 13
-
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline
Paper • 2404.02893 • Published • 21 -
MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
Paper • 2309.12284 • Published • 19 -
Premise Order Matters in Reasoning with Large Language Models
Paper • 2402.08939 • Published • 28
-
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
Locating and Editing Factual Associations in Mamba
Paper • 2404.03646 • Published • 3 -
Locating and Editing Factual Associations in GPT
Paper • 2202.05262 • Published • 1 -
KAN: Kolmogorov-Arnold Networks
Paper • 2404.19756 • Published • 116
-
HyperCLOVA X Technical Report
Paper • 2404.01954 • Published • 25 -
UltraFeedback: Boosting Language Models with High-quality Feedback
Paper • 2310.01377 • Published • 6 -
AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback
Paper • 2305.14387 • Published • 1 -
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Paper • 2402.03300 • Published • 145
-
On the Scalability of Diffusion-based Text-to-Image Generation
Paper • 2404.02883 • Published • 19 -
MonoPatchNeRF: Improving Neural Radiance Fields with Patch-based Monocular Guidance
Paper • 2404.08252 • Published • 6 -
Scaling Down to Scale Up: A Guide to Parameter-Efficient Fine-Tuning
Paper • 2303.15647 • Published • 4
-
Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction
Paper • 2404.02905 • Published • 74 -
ControlNet++: Improving Conditional Controls with Efficient Consistency Feedback
Paper • 2404.07987 • Published • 48 -
COCONut: Modernizing COCO Segmentation
Paper • 2404.08639 • Published • 30 -
MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs
Paper • 2402.15627 • Published • 36
-
Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction
Paper • 2404.02905 • Published • 74 -
On Speculative Decoding for Multimodal Large Language Models
Paper • 2404.08856 • Published • 13 -
Hydragen: High-Throughput LLM Inference with Shared Prefixes
Paper • 2402.05099 • Published • 20
-
Freditor: High-Fidelity and Transferable NeRF Editing by Frequency Decomposition
Paper • 2404.02514 • Published • 11 -
MonoPatchNeRF: Improving Neural Radiance Fields with Patch-based Monocular Guidance
Paper • 2404.08252 • Published • 6 -
Video2Game: Real-time, Interactive, Realistic and Browser-Compatible Environment from a Single Video
Paper • 2404.09833 • Published • 30 -
MeshLRM: Large Reconstruction Model for High-Quality Mesh
Paper • 2404.12385 • Published • 27
-
Freditor: High-Fidelity and Transferable NeRF Editing by Frequency Decomposition
Paper • 2404.02514 • Published • 11 -
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model
Paper • 2404.04167 • Published • 13 -
Length Generalization of Causal Transformers without Position Encoding
Paper • 2404.12224 • Published • 1 -
Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
Paper • 2406.07394 • Published • 29
-
I-Design: Personalized LLM Interior Designer
Paper • 2404.02838 • Published • 2 -
Scaling MLPs: A Tale of Inductive Bias
Paper • 2306.13575 • Published • 17 -
Fast Feedforward Networks
Paper • 2308.14711 • Published • 3 -
How Good Are Low-bit Quantized LLaMA3 Models? An Empirical Study
Paper • 2404.14047 • Published • 45
-
I am a Strange Dataset: Metalinguistic Tests for Language Models
Paper • 2401.05300 • Published • 3 -
Megalodon: Efficient LLM Pretraining and Inference with Unlimited Context Length
Paper • 2404.08801 • Published • 66 -
SpecInfer: Accelerating Generative LLM Serving with Speculative Inference and Token Tree Verification
Paper • 2305.09781 • Published • 4 -
MeshLRM: Large Reconstruction Model for High-Quality Mesh
Paper • 2404.12385 • Published • 27
-
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
Premise Order Matters in Reasoning with Large Language Models
Paper • 2402.08939 • Published • 28 -
Reasoning in Large Language Models: A Geometric Perspective
Paper • 2407.02678 • Published • 1 -
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Paper • 2402.03300 • Published • 145
-
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
Paper • 2406.07394 • Published • 29 -
Qwen2 Technical Report
Paper • 2407.10671 • Published • 172 -
Self-Consistency Preference Optimization
Paper • 2411.04109 • Published • 19
Repo: https://github.com/HistLab/More-efficient-manual-review-of-automatically-transcribed-tabular-data
-
CascadeTabNet: An approach for end to end table detection and structure recognition from image-based documents
Paper • 2004.12629 • Published • 3 -
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking
Paper • 2204.08387 • Published • 8 -
Text Role Classification in Scientific Charts Using Multimodal Transformers
Paper • 2402.14579 • Published • 1 -
An inclusive review on deep learning techniques and their scope in handwriting recognition
Paper • 2404.08011 • Published • 1
-
LayoutLM: Pre-training of Text and Layout for Document Image Understanding
Paper • 1912.13318 • Published • 5 -
LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding
Paper • 2012.14740 • Published • 3 -
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking
Paper • 2204.08387 • Published • 8
-
CodeEditorBench: Evaluating Code Editing Capability of Large Language Models
Paper • 2404.03543 • Published • 18 -
McEval: Massively Multilingual Code Evaluation
Paper • 2406.07436 • Published • 41 -
BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions
Paper • 2406.15877 • Published • 50 -
Qwen2 Technical Report
Paper • 2407.10671 • Published • 172
MAWPS paper: https://aclanthology.org/N16-1136.pdf
-
Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problems
Paper • 1705.04146 • Published • 1 -
Training Verifiers to Solve Math Word Problems
Paper • 2110.14168 • Published • 7 -
Explaining Math Word Problem Solvers
Paper • 2307.13128 • Published • 1 -
MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms
Paper • 1905.13319 • Published • 2
-
Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problems
Paper • 1705.04146 • Published • 1 -
Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?
Paper • 2404.03411 • Published • 10 -
No "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance
Paper • 2404.04125 • Published • 29 -
Hydragen: High-Throughput LLM Inference with Shared Prefixes
Paper • 2402.05099 • Published • 20
Alpaca eval: https://github.com/tatsu-lab/alpaca_eval
-
UltraFeedback: Boosting Language Models with High-quality Feedback
Paper • 2310.01377 • Published • 6 -
Learn Your Reference Model for Real Good Alignment
Paper • 2404.09656 • Published • 90 -
Natural Language Reinforcement Learning
Paper • 2411.14251 • Published • 31 -
Group Robust Preference Optimization in Reward-free RLHF
Paper • 2405.20304 • Published • 1
-
Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?
Paper • 2404.03411 • Published • 10 -
The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions
Paper • 2404.13208 • Published • 40 -
A False Sense of Safety: Unsafe Information Leakage in 'Safe' AI Responses
Paper • 2407.02551 • Published • 9
-
CohereLabs/c4ai-command-r-plus
Text Generation • Updated • 2.81k • 1.78k -
CohereLabs/c4ai-command-r-plus-4bit
Text Generation • 105B • Updated • 37 • 258 -
mistral-community/Mixtral-8x22B-v0.1-4bit
Text Generation • 145B • Updated • 17 • 55 -
CohereLabs/c4ai-command-r-v01
Text Generation • Updated • 22.4k • 1.11k
-
openai/whisper-large-v3
Automatic Speech Recognition • 2B • Updated • 4.94M • • 5.68k -
openai/whisper-tiny
Automatic Speech Recognition • Updated • 768k • 427 -
openai/whisper-small
Automatic Speech Recognition • Updated • 2.28M • 558 -
openai/whisper-medium
Automatic Speech Recognition • Updated • 909k • 284
OpenAI has a 2024 tool referring to this technique: https://github.com/openai/transformer-debugger with https://transformer-circuits.pub/2023/monosema
Websit: https://rome.baulab.info/
-
Analyzing Transformers in Embedding Space
Paper • 2209.02535 • Published • 3 -
Prompt-to-Prompt Image Editing with Cross Attention Control
Paper • 2208.01626 • Published • 3 -
Dynamic Typography: Bringing Words to Life
Paper • 2404.11614 • Published • 46 -
Transformer Language Models without Positional Encodings Still Learn Positional Information
Paper • 2203.16634 • Published • 6
https://github.com/jessevig/bertviz
FFNs: Using a smoother GELU instead of an ReLu
A batched on-policy algorithm that conducts self-improvement iteratively via contrastive learning
a multi-step Markov Decision Process, allowing one to fine-tune consistency models toward a downstream task using just a reward function.
Focuses on policy improvement through search-based sampling
-
MLP Can Be A Good Transformer Learner
Paper • 2404.05657 • Published • 1 -
Toward a Better Understanding of Fourier Neural Operators: Analysis and Improvement from a Spectral Perspective
Paper • 2404.07200 • Published • 2 -
An inclusive review on deep learning techniques and their scope in handwriting recognition
Paper • 2404.08011 • Published • 1 -
Long-form music generation with latent diffusion
Paper • 2404.10301 • Published • 27
-
Kandinsky: an Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion
Paper • 2310.03502 • Published • 79 -
Transferable and Principled Efficiency for Open-Vocabulary Segmentation
Paper • 2404.07448 • Published • 12 -
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32 -
COCONut: Modernizing COCO Segmentation
Paper • 2404.08639 • Published • 30
https://machinelearningmastery.com/how-to-implement-the-frechet-inception-distance-fid-from-scratch/
-
Kandinsky: an Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion
Paper • 2310.03502 • Published • 79 -
GLIGEN: Open-Set Grounded Text-to-Image Generation
Paper • 2301.07093 • Published • 4 -
Music Consistency Models
Paper • 2404.13358 • Published • 14 -
Align Your Steps: Optimizing Sampling Schedules in Diffusion Models
Paper • 2404.14507 • Published • 23
Like coupled masses connected by springs
-
ConsistencyDet: Robust Object Detector with Denoising Paradigm of Consistency Model
Paper • 2404.07773 • Published • 1 -
Hyper-SD: Trajectory Segmented Consistency Model for Efficient Image Synthesis
Paper • 2404.13686 • Published • 29 -
Align Your Steps: Optimizing Sampling Schedules in Diffusion Models
Paper • 2404.14507 • Published • 23 -
NAF-DPM: A Nonlinear Activation-Free Diffusion Probabilistic Model for Document Enhancement
Paper • 2404.05669 • Published • 1
-
facebook/maskformer-swin-base-coco
Image Segmentation • 0.1B • Updated • 863 • • 26 -
facebook/mask2former-swin-small-coco-panoptic
Image Segmentation • 68.8M • Updated • 1.52k • 1 -
facebook/mask2former-swin-tiny-coco-instance
Image Segmentation • 47.5M • Updated • 57k • • 15 -
facebook/mask2former-swin-small-ade-semantic
Image Segmentation • 68.8M • Updated • 22.5k • • 8
Audio reading using bark: https://drive.google.com/file/d/1AlHLzeUd04LXgDj99SOvmQJTy9chufGo/view?usp=sharing
-
ODA: Observation-Driven Agent for integrating LLMs and Knowledge Graphs
Paper • 2404.07677 • Published • 1 -
ResearchAgent: Iterative Research Idea Generation over Scientific Literature with Large Language Models
Paper • 2404.07738 • Published • 2 -
Scaling Instructable Agents Across Many Simulated Worlds
Paper • 2404.10179 • Published • 28 -
A Multimodal Automated Interpretability Agent
Paper • 2404.14394 • Published • 23
-
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32 -
GLIGEN: Open-Set Grounded Text-to-Image Generation
Paper • 2301.07093 • Published • 4 -
PhysDreamer: Physics-Based Interaction with 3D Objects via Video Generation
Paper • 2404.13026 • Published • 24 -
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15
-
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32 -
COCONut: Modernizing COCO Segmentation
Paper • 2404.08639 • Published • 30 -
GLIGEN: Open-Set Grounded Text-to-Image Generation
Paper • 2301.07093 • Published • 4 -
Grounded Language-Image Pre-training
Paper • 2112.03857 • Published • 3
Read by Bark: https://drive.google.com/file/d/1qltkY31-013JDQn-u2pmnjPyCaUcOqsV/view?usp=sharing
Read by Bark: https://drive.google.com/file/d/1apmyvLMEQ97ObHKzQna9URFHF0Xg-EsO/view?usp=sharing
Read by Bark: https://drive.google.com/file/d/1AlHLzeUd04LXgDj99SOvmQJTy9chufGo/view?usp=sharing
-
Instruction Tuning with Human Curriculum
Paper • 2310.09518 • Published • 3 -
A Thorough Examination of Decoding Methods in the Era of LLMs
Paper • 2402.06925 • Published • 1 -
Distilling System 2 into System 1
Paper • 2407.06023 • Published • 4 -
Byte Latent Transformer: Patches Scale Better Than Tokens
Paper • 2412.09871 • Published • 108
Reading by Bark: https://drive.google.com/file/d/1cdKmflJ3jRKszi5s3RF4Ru6nUw8tvpHX/view?usp=sharing reading duration 3 hours and 28 minutes
Read by Bark: https://drive.google.com/file/d/1fEZ8uwnfniMljZ5S60NxOav6Qav2A-XB/view?usp=sharing duration is 44m 12s
The DAC decoder tanh caused harmonic distortion
an encoder-decoder model
which compresses videos to discrete embeddings (tokens) and a transformer model to translate
text embeddings to video tokens.
Fig 10. CFG substantially improves language conditionality.
MaskGiT is trained to reconstruct
masked tokens z predicted by a frozen C-ViViT encoder and conditioned on T5X tokens of a given
prompt p0
-
On the Scalability of GNNs for Molecular Graphs
Paper • 2404.11568 • Published • 1 -
Octopus v4: Graph of language models
Paper • 2404.19296 • Published • 118 -
Architectures of Topological Deep Learning: A Survey on Topological Neural Networks
Paper • 2304.10031 • Published • 3 -
Meta Flow Matching: Integrating Vector Fields on the Wasserstein Manifold
Paper • 2408.14608 • Published • 8
1) random walk diagonals 2) Laplacian eigenvectors for geometry and position 3) global structural information about the graph
-
Explainable Lung Disease Classification from Chest X-Ray Images Utilizing Deep Learning and XAI
Paper • 2404.11428 • Published • 1 -
A Multimodal Automated Interpretability Agent
Paper • 2404.14394 • Published • 23 -
What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation
Paper • 2404.07129 • Published • 3 -
The Geometry of Categorical and Hierarchical Concepts in Large Language Models
Paper • 2406.01506 • Published • 3
LIME
-
MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs
Paper • 2402.15627 • Published • 36 -
One Wide Feedforward is All You Need
Paper • 2309.01826 • Published • 34 -
Fast Feedforward Networks
Paper • 2308.14711 • Published • 3 -
Memory Layers at Scale
Paper • 2412.09764 • Published • 5
See Page 7 Figure 5 on right: Repeating for 4 epochs is almost as good as new data
contains documents and summaries in six languages (German, English, Spanish, Russian, Turkish, and Vietnamese) with pointwise human ratings
-
TriForce: Lossless Acceleration of Long Sequence Generation with Hierarchical Speculative Decoding
Paper • 2404.11912 • Published • 17 -
SnapKV: LLM Knows What You are Looking for Before Generation
Paper • 2404.14469 • Published • 27 -
LLM in a flash: Efficient Large Language Model Inference with Limited Memory
Paper • 2312.11514 • Published • 264 -
An Evolved Universal Transformer Memory
Paper • 2410.13166 • Published • 6
we utilize a 4K retrieval cache as an intermediate draft cache in our hierarchical system, while leveraging the JackFram/Llama68M (JF68M) [28] model
-
MeshLRM: Large Reconstruction Model for High-Quality Mesh
Paper • 2404.12385 • Published • 27 -
MaPa: Text-driven Photorealistic Material Painting for 3D Shapes
Paper • 2404.17569 • Published • 13 -
MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers
Paper • 2406.10163 • Published • 33 -
Meta 3D AssetGen: Text-to-Mesh Generation with High-Quality Geometry, Texture, and PBR Materials
Paper • 2407.02445 • Published • 4
Uniform scaling not as good as Head-based scaling
Page 7 fig shows NoPE extending passed the models context size from pretraining or fine-tuning
-
The Impact of Positional Encoding on Length Generalization in Transformers
Paper • 2305.19466 • Published • 2 -
Transformers Can Do Arithmetic with the Right Embeddings
Paper • 2405.17399 • Published • 54 -
Teaching Transformers Causal Reasoning through Axiomatic Training
Paper • 2407.07612 • Published • 2 -
Round and Round We Go! What makes Rotary Positional Encodings useful?
Paper • 2410.06205 • Published • 2
-
RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval
Paper • 2401.18059 • Published • 48 -
KGValidator: A Framework for Automatic Validation of Knowledge Graph Construction
Paper • 2404.15923 • Published • 2 -
The Geometry of Categorical and Hierarchical Concepts in Large Language Models
Paper • 2406.01506 • Published • 3
-
GLIGEN: Open-Set Grounded Text-to-Image Generation
Paper • 2301.07093 • Published • 4 -
YOLO-World: Real-Time Open-Vocabulary Object Detection
Paper • 2401.17270 • Published • 43 -
DETRs Beat YOLOs on Real-time Object Detection
Paper • 2304.08069 • Published • 16 -
RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformer
Paper • 2407.17140 • Published • 2
Driving particles
-
Hyper-SD: Trajectory Segmented Consistency Model for Efficient Image Synthesis
Paper • 2404.13686 • Published • 29 -
MultiBooth: Towards Generating All Your Concepts in an Image from Text
Paper • 2404.14239 • Published • 9 -
Stylus: Automatic Adapter Selection for Diffusion Models
Paper • 2404.18928 • Published • 15 -
MagicQuill: An Intelligent Interactive Image Editing System
Paper • 2411.09703 • Published • 80
Multi-modal Concept Extraction
-
All you need is a good init
Paper • 1511.06422 • Published • 1 -
Align Your Steps: Optimizing Sampling Schedules in Diffusion Models
Paper • 2404.14507 • Published • 23 -
Deep Residual Learning for Image Recognition
Paper • 1512.03385 • Published • 16 -
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15
-
OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework
Paper • 2404.14619 • Published • 126 -
Scaling Down to Scale Up: A Guide to Parameter-Efficient Fine-Tuning
Paper • 2303.15647 • Published • 4 -
Hyper-X: A Unified Hypernetwork for Multi-Task Multilingual Transfer
Paper • 2205.12148 • Published • 2 -
No More Adam: Learning Rate Scaling at Initialization is All You Need
Paper • 2412.11768 • Published • 43
-
OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework
Paper • 2404.14619 • Published • 126 -
apple/OpenELM-450M
Text Generation • 0.5B • Updated • 652 • 26 -
apple/OpenELM-3B
Text Generation • 3B • Updated • 401 • 130 -
apple/OpenELM-3B-Instruct
Text Generation • 3B • Updated • 3.24k • 339
https://stanfordnlp.github.io/CoreNLP/openie.html
https://wordnet.princeton.edu/
Using SimCSE
-
LEGENT: Open Platform for Embodied Agents
Paper • 2404.18243 • Published • 22 -
Ag2Manip: Learning Novel Manipulation Skills with Agent-Agnostic Visual and Action Representations
Paper • 2404.17521 • Published • 13 -
Octopus v4: Graph of language models
Paper • 2404.19296 • Published • 118 -
AgentOhana: Design Unified Data and Training Pipeline for Effective Agent Learning
Paper • 2402.15506 • Published • 18
Two layer induction heads
-
What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation
Paper • 2404.07129 • Published • 3 -
In-context Learning and Induction Heads
Paper • 2209.11895 • Published • 2 -
pyvene: A Library for Understanding and Improving PyTorch Models via Interventions
Paper • 2403.07809 • Published • 1
Additional reading: https://transformer-circuits.pub/2021/framework/index.html
-
In-context Learning and Induction Heads
Paper • 2209.11895 • Published • 2 -
What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation
Paper • 2404.07129 • Published • 3 -
pyvene: A Library for Understanding and Improving PyTorch Models via Interventions
Paper • 2403.07809 • Published • 1
Modifying activations during training with proper gradient flow
Larger vocab is better compression, but may result in longer training ICL phase change delays due to the slower Induction Head Copy Subcircuit (C)
-
Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
Paper • 2405.21060 • Published • 68 -
Chimera: Effectively Modeling Multivariate Time Series with 2-Dimensional State Space Models
Paper • 2406.04320 • Published • 10 -
MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection
Paper • 2403.19888 • Published • 12 -
Graph Mamba: Towards Learning on Graphs with State Space Models
Paper • 2402.08678 • Published • 17
Section 7.3.3
Table 7
Guidance - "The intended effect is to decrease the diversity of the samples while increasing the quality of each individual sample."
-
Classifier-Free Diffusion Guidance
Paper • 2207.12598 • Published • 5 -
Adding Conditional Control to Text-to-Image Diffusion Models
Paper • 2302.05543 • Published • 58 -
Applying Guidance in a Limited Interval Improves Sample and Distribution Quality in Diffusion Models
Paper • 2404.07724 • Published • 14 -
Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation
Paper • 2406.06525 • Published • 71
-
DataComp: In search of the next generation of multimodal datasets
Paper • 2304.14108 • Published • 2 -
The Synergy between Data and Multi-Modal Large Language Models: A Survey from Co-Development Perspective
Paper • 2407.08583 • Published • 13 -
TIP-I2V: A Million-Scale Real Text and Image Prompt Dataset for Image-to-Video Generation
Paper • 2411.04709 • Published • 27 -
YFCC100M: The New Data in Multimedia Research
Paper • 1503.01817 • Published • 1
-
CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs
Paper • 2406.18521 • Published • 31 -
We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?
Paper • 2407.01284 • Published • 81 -
ChartGemma: Visual Instruction-tuning for Chart Reasoning in the Wild
Paper • 2407.04172 • Published • 25
-
HuatuoGPT-Vision, Towards Injecting Medical Visual Knowledge into Multimodal LLMs at Scale
Paper • 2406.19280 • Published • 63 -
PPLLaVA: Varied Video Sequence Understanding With Prompt Guidance
Paper • 2411.02327 • Published • 11 -
MagicQuill: An Intelligent Interactive Image Editing System
Paper • 2411.09703 • Published • 80 -
LLaVA-o1: Let Vision Language Models Reason Step-by-Step
Paper • 2411.10440 • Published • 131
Repo: https://github.com/EleutherAI/sae
-
nomic-ai/nomic-embed-text-v1
Sentence Similarity • 0.1B • Updated • 7.33M • 567 -
nomic-ai/nomic-embed-text-v1.5-GGUF
Sentence Similarity • 0.1B • Updated • 90.3k • 103 -
BAAI/bge-large-en-v1.5
Feature Extraction • 0.3B • Updated • 15.4M • • 662 -
mixedbread-ai/mxbai-embed-2d-large-v1
Feature Extraction • 0.3B • Updated • 149k • • 43
Repo: https://github.com/shengliu66/ICV
See also: https://dawn.cs.stanford.edu/2019/10/10/noneuclidean/
See also: https://dawn.cs.stanford.edu/2019/10/10/noneuclidean/
-
In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering
Paper • 2311.06668 • Published • 5 -
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Efficient Algorithms for t-distributed Stochastic Neighborhood Embedding
Paper • 1712.09005 • Published • 1
Recommended to explore constrained manifold areas with limited curvature (possible bias)
See also: "A Riemannian Framework for Tensor Computing" https://inria.hal.science/inria-00070743/file/RR-5255.pdf
-
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Barycentric Subspace Analysis on Manifolds
Paper • 1607.02833 • Published • 1 -
Poincaré Embeddings for Learning Hierarchical Representations
Paper • 1705.08039 • Published • 1 -
Efficient Algorithms for t-distributed Stochastic Neighborhood Embedding
Paper • 1712.09005 • Published • 1
Also see: https://dawn.cs.stanford.edu/2019/10/10/noneuclidean/
-
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Architectures of Topological Deep Learning: A Survey on Topological Neural Networks
Paper • 2304.10031 • Published • 3 -
Galois Theory
Paper • 2408.07499 • Published • 1 -
Equivariant Transformer Networks
Paper • 1901.11399 • Published • 1
The manifold needs to be known for this layer to be implemented, and manifolds whose Exp enjoys an analytical expression are preferred
Euclidean signal on Euclidean domains for keys, queries and values, with group action on codomain.
Manifold-valued data convolutions. Tangent mean can be used for Fréchet mean to save on compute
Repo: https://github.com/TorchJD/torchjd
Helps with finding the end of the byte patch
-
ByT5: Towards a token-free future with pre-trained byte-to-byte models
Paper • 2105.13626 • Published • 5 -
Byte Latent Transformer: Patches Scale Better Than Tokens
Paper • 2412.09871 • Published • 108 -
MrT5: Dynamic Token Merging for Efficient Byte-level Language Models
Paper • 2410.20771 • Published • 3
More extracted images on github: https://github.com/matlok-ai/python-copilot-image-and-audio-examples/tree/main/png
-
matlok/python-image-copilot-training-using-class-knowledge-graphs-2024-01-27
Viewer • Updated • 773 • 326 -
matlok/python-image-copilot-training-using-function-knowledge-graphs
Viewer • Updated • 88 • 1.41k -
matlok/python-image-copilot-training-using-inheritance-knowledge-graphs
Viewer • Updated • 88 • 1.38k -
matlok/python-image-copilot-training-using-import-knowledge-graphs
Viewer • Updated • 88 • 229
More extracted mp3 samples on github: https://github.com/matlok-ai/python-copilot-image-and-audio-examples/tree/main/mp3
-
matlok/python-audio-copilot-training-using-class-knowledge-graphs-2024-01-27
Viewer • Updated • 948 • 567 -
matlok/python-audio-copilot-training-using-function-knowledge-graphs
Viewer • Updated • 120 • 1.92k • 1 -
matlok/python-audio-copilot-training-using-inheritance-knowledge-graphs
Viewer • Updated • 120 • 654 -
matlok/python-audio-copilot-training-using-import-knowledge-graphs
Viewer • Updated • 48 • 410
Great article on how this works: https://towardsdatascience.com/a-beginners-guide-to-llm-fine-tuning-4bae7d4da672
-
Visual Instruction Tuning
Paper • 2304.08485 • Published • 21 -
LLaVA-Plus: Learning to Use Tools for Creating Multimodal Agents
Paper • 2311.05437 • Published • 51 -
Improved Baselines with Visual Instruction Tuning
Paper • 2310.03744 • Published • 39 -
Aligning Large Multimodal Models with Factually Augmented RLHF
Paper • 2309.14525 • Published • 32
-
Self-Instruct: Aligning Language Model with Self Generated Instructions
Paper • 2212.10560 • Published • 9 -
Principled Instructions Are All You Need for Questioning LLaMA-1/2, GPT-3.5/4
Paper • 2312.16171 • Published • 37 -
DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence
Paper • 2401.14196 • Published • 72 -
AlpaCare:Instruction-tuned Large Language Models for Medical Application
Paper • 2310.14558 • Published • 4
MoE
-
MoE-LLaVA: Mixture of Experts for Large Vision-Language Models
Paper • 2401.15947 • Published • 53 -
DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models
Paper • 2401.06066 • Published • 61 -
SwitchHead: Accelerating Transformers with Mixture-of-Experts Attention
Paper • 2312.07987 • Published • 41 -
Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity
Paper • 2101.03961 • Published • 13
-
dphn/dolphin-2.6-mistral-7b-dpo-laser
Text Generation • 7B • Updated • 77 • 120 -
Evaluating Large Language Models Trained on Code
Paper • 2107.03374 • Published • 10 -
CodeBERT: A Pre-Trained Model for Programming and Natural Languages
Paper • 2002.08155 • Published • 2 -
code2seq: Generating Sequences from Structured Representations of Code
Paper • 1808.01400 • Published • 2
Checkout: https://bbycroft.net/llm and http://nlp.seas.harvard.edu/2018/04/03/attention.html
-
Attention Is All You Need
Paper • 1706.03762 • Published • 122 -
ImageNet Large Scale Visual Recognition Challenge
Paper • 1409.0575 • Published • 10 -
Sequence to Sequence Learning with Neural Networks
Paper • 1409.3215 • Published • 3 -
Language Models are Few-Shot Learners
Paper • 2005.14165 • Published • 20
-
LCM-LoRA: A Universal Stable-Diffusion Acceleration Module
Paper • 2311.05556 • Published • 86 -
MultiLoRA: Democratizing LoRA for Better Multi-Task Learning
Paper • 2311.11501 • Published • 37 -
S-LoRA: Serving Thousands of Concurrent LoRA Adapters
Paper • 2311.03285 • Published • 30 -
LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70B
Paper • 2310.20624 • Published • 13
-
Metadata Might Make Language Models Better
Paper • 2211.10086 • Published • 4 -
Empirical Analysis of the Strengths and Weaknesses of PEFT Techniques for LLMs
Paper • 2304.14999 • Published • 2 -
PEFT for Speech: Unveiling Optimal Placement, Merging Strategies, and Ensemble Techniques
Paper • 2401.02122 • Published • 2 -
Zephyr: Direct Distillation of LM Alignment
Paper • 2310.16944 • Published • 123
-
Spam-T5: Benchmarking Large Language Models for Few-Shot Email Spam Detection
Paper • 2304.01238 • Published • 2 -
The FinBen: An Holistic Financial Benchmark for Large Language Models
Paper • 2402.12659 • Published • 24 -
TofuEval: Evaluating Hallucinations of LLMs on Topic-Focused Dialogue Summarization
Paper • 2402.13249 • Published • 15 -
Evaluating Large Language Models Trained on Code
Paper • 2107.03374 • Published • 10
U-Net was trained in 10 hours on a NVidia Titan GPU (6 GB) - 2015
-
U-Net: Convolutional Networks for Biomedical Image Segmentation
Paper • 1505.04597 • Published • 18 -
microsoft/BioGPT-Large
Text Generation • Updated • 5.44k • 238 -
kuleshov-group/caduceus-ps_seqlen-131k_d_model-256_n_layer-16
Fill-Mask • 7.73M • Updated • 2.02k • 14 -
kuleshov-group/caduceus-ps_seqlen-1k_d_model-256_n_layer-4_lr-8e-3
Fill-Mask • 1.93M • Updated • 527 • 2
-
LAVE: LLM-Powered Agent Assistance and Language Augmentation for Video Editing
Paper • 2402.10294 • Published • 27 -
Valley: Video Assistant with Large Language model Enhanced abilitY
Paper • 2306.07207 • Published • 3 -
Video Editing via Factorized Diffusion Distillation
Paper • 2403.09334 • Published • 22
-
Linear Transformers with Learnable Kernel Functions are Better In-Context Models
Paper • 2402.10644 • Published • 81 -
GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
Paper • 2305.13245 • Published • 6 -
ChunkAttention: Efficient Self-Attention with Prefix-Aware KV Cache and Two-Phase Partition
Paper • 2402.15220 • Published • 20 -
Sequence Parallelism: Long Sequence Training from System Perspective
Paper • 2105.13120 • Published • 6
-
DataDreamer: A Tool for Synthetic Data Generation and Reproducible LLM Workflows
Paper • 2402.10379 • Published • 31 -
Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Grasping
Paper • 1709.07857 • Published • 2 -
Simple synthetic data reduces sycophancy in large language models
Paper • 2308.03958 • Published • 23 -
Qwen-VL: A Frontier Large Vision-Language Model with Versatile Abilities
Paper • 2308.12966 • Published • 12
-
FinTral: A Family of GPT-4 Level Multimodal Financial Large Language Models
Paper • 2402.10986 • Published • 83 -
BloombergGPT: A Large Language Model for Finance
Paper • 2303.17564 • Published • 32 -
GPT-InvestAR: Enhancing Stock Investment Strategies through Annual Report Analysis with Large Language Models
Paper • 2309.03079 • Published • 3 -
FinVis-GPT: A Multimodal Large Language Model for Financial Chart Analysis
Paper • 2308.01430 • Published • 2
-
Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Grasping
Paper • 1709.07857 • Published • 2 -
Sensor-based Multi-Robot Search and Coverage with Spatial Separation in Unstructured Environments
Paper • 2403.01710 • Published • 2 -
Twisting Lids Off with Two Hands
Paper • 2403.02338 • Published • 7
exploring speculative sampling with autoregressive model like: https://proceedings.mlr.press/v139/song21a.html and https://proceedings.mlr.press/v119/
-
CodeBERT: A Pre-Trained Model for Programming and Natural Languages
Paper • 2002.08155 • Published • 2 -
OpenCodeInterpreter: Integrating Code Generation with Execution and Refinement
Paper • 2402.14658 • Published • 84 -
CodeFusion: A Pre-trained Diffusion Model for Code Generation
Paper • 2310.17680 • Published • 74 -
CodePlan: Repository-level Coding using LLMs and Planning
Paper • 2309.12499 • Published • 80
-
Orca-Math: Unlocking the potential of SLMs in Grade School Math
Paper • 2402.14830 • Published • 24 -
MathScale: Scaling Instruction Tuning for Mathematical Reasoning
Paper • 2403.02884 • Published • 17 -
meta-math/MetaMath-Mistral-7B
Text Generation • Updated • 1.26k • • 96 -
meta-math/MetaMath-13B-V1.0
Text Generation • Updated • 1k • 13
-
Same Task, More Tokens: the Impact of Input Length on the Reasoning Performance of Large Language Models
Paper • 2402.14848 • Published • 19 -
Teaching Large Language Models to Reason with Reinforcement Learning
Paper • 2403.04642 • Published • 48 -
How Far Are We from Intelligent Visual Deductive Reasoning?
Paper • 2403.04732 • Published • 21 -
Learning to Reason and Memorize with Self-Notes
Paper • 2305.00833 • Published • 5
-
MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases
Paper • 2402.14905 • Published • 134 -
Sensor-based Multi-Robot Search and Coverage with Spatial Separation in Unstructured Environments
Paper • 2403.01710 • Published • 2 -
EdgeMoE: Fast On-Device Inference of MoE-based Large Language Models
Paper • 2308.14352 • Published -
Slimmable Encoders for Flexible Split DNNs in Bandwidth and Resource Constrained IoT Systems
Paper • 2306.12691 • Published • 3
-
Video as the New Language for Real-World Decision Making
Paper • 2402.17139 • Published • 22 -
VideoCrafter1: Open Diffusion Models for High-Quality Video Generation
Paper • 2310.19512 • Published • 16 -
VideoMamba: State Space Model for Efficient Video Understanding
Paper • 2403.06977 • Published • 29 -
VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models
Paper • 2401.09047 • Published • 14
-
introspector/unimath
Updated • 3.04k • 7 -
MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms
Paper • 1905.13319 • Published • 2 -
Measuring Mathematical Problem Solving With the MATH Dataset
Paper • 2103.03874 • Published • 6 -
MathScale: Scaling Instruction Tuning for Mathematical Reasoning
Paper • 2403.02884 • Published • 17
-
Priority Sampling of Large Language Models for Compilers
Paper • 2402.18734 • Published • 19 -
Accelerating Large Language Model Decoding with Speculative Sampling
Paper • 2302.01318 • Published • 4 -
Fast Inference from Transformers via Speculative Decoding
Paper • 2211.17192 • Published • 11 -
AttentiveNAS: Improving Neural Architecture Search via Attentive Sampling
Paper • 2011.09011 • Published • 2
-
Training Verifiers to Solve Math Word Problems
Paper • 2110.14168 • Published • 7 -
MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
Paper • 2309.12284 • Published • 19 -
LiteSearch: Efficacious Tree Search for LLM
Paper • 2407.00320 • Published • 40 -
DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models
Paper • 2309.03883 • Published • 36
-
Measuring the Effects of Data Parallelism on Neural Network Training
Paper • 1811.03600 • Published • 2 -
Adafactor: Adaptive Learning Rates with Sublinear Memory Cost
Paper • 1804.04235 • Published • 2 -
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
Paper • 1905.11946 • Published • 4 -
Yi: Open Foundation Models by 01.AI
Paper • 2403.04652 • Published • 65
-
Sequence Parallelism: Long Sequence Training from System Perspective
Paper • 2105.13120 • Published • 6 -
Ring Attention with Blockwise Transformers for Near-Infinite Context
Paper • 2310.01889 • Published • 13 -
Striped Attention: Faster Ring Attention for Causal Transformers
Paper • 2311.09431 • Published • 4 -
World Model on Million-Length Video And Language With RingAttention
Paper • 2402.08268 • Published • 40
-
Accelerating LLM Inference with Staged Speculative Decoding
Paper • 2308.04623 • Published • 26 -
An Emulator for Fine-Tuning Large Language Models using Small Language Models
Paper • 2310.12962 • Published • 13 -
The Curious Case of Neural Text Degeneration
Paper • 1904.09751 • Published • 3 -
On Speculative Decoding for Multimodal Large Language Models
Paper • 2404.08856 • Published • 13
-
ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment
Paper • 2403.05135 • Published • 45 -
DeepSeek-VL: Towards Real-World Vision-Language Understanding
Paper • 2403.05525 • Published • 49 -
CoCa: Contrastive Captioners are Image-Text Foundation Models
Paper • 2205.01917 • Published • 3
-
Mixtral of Experts
Paper • 2401.04088 • Published • 160 -
MoE-LLaVA: Mixture of Experts for Large Vision-Language Models
Paper • 2401.15947 • Published • 53 -
MoE-Mamba: Efficient Selective State Space Models with Mixture of Experts
Paper • 2401.04081 • Published • 74 -
EdgeMoE: Fast On-Device Inference of MoE-based Large Language Models
Paper • 2308.14352 • Published
-
Adaptive sequential Monte Carlo by means of mixture of experts
Paper • 1108.2836 • Published • 2 -
Convergence Rates for Mixture-of-Experts
Paper • 1110.2058 • Published • 2 -
Multi-view Contrastive Learning for Entity Typing over Knowledge Graphs
Paper • 2310.12008 • Published • 2 -
Enhancing NeRF akin to Enhancing LLMs: Generalizable NeRF Transformer with Mixture-of-View-Experts
Paper • 2308.11793 • Published • 2
-
Robust Mixture-of-Expert Training for Convolutional Neural Networks
Paper • 2308.10110 • Published • 2 -
Experts Weights Averaging: A New General Training Scheme for Vision Transformers
Paper • 2308.06093 • Published • 2 -
ConstitutionalExperts: Training a Mixture of Principle-based Prompts
Paper • 2403.04894 • Published • 2 -
Mixture-of-LoRAs: An Efficient Multitask Tuning for Large Language Models
Paper • 2403.03432 • Published • 1
-
Mixture-of-LoRAs: An Efficient Multitask Tuning for Large Language Models
Paper • 2403.03432 • Published • 1 -
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data
Paper • 2304.08247 • Published • 2
-
Turn Waste into Worth: Rectifying Top-k Router of MoE
Paper • 2402.12399 • Published • 2 -
CompeteSMoE -- Effective Training of Sparse Mixture of Experts via Competition
Paper • 2402.02526 • Published • 3 -
Buffer Overflow in Mixture of Experts
Paper • 2402.05526 • Published • 9 -
OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models
Paper • 2402.01739 • Published • 28
-
Scaling Vision with Sparse Mixture of Experts
Paper • 2106.05974 • Published • 4 -
Routers in Vision Mixture of Experts: An Empirical Study
Paper • 2401.15969 • Published • 2 -
Multimodal Contrastive Learning with LIMoE: the Language-Image Mixture of Experts
Paper • 2206.02770 • Published • 4 -
Experts Weights Averaging: A New General Training Scheme for Vision Transformers
Paper • 2308.06093 • Published • 2
-
LocMoE: A Low-overhead MoE for Large Language Model Training
Paper • 2401.13920 • Published • 2 -
HyperRouter: Towards Efficient Training and Inference of Sparse Mixture of Experts
Paper • 2312.07035 • Published • 2 -
DEMix Layers: Disentangling Domains for Modular Language Modeling
Paper • 2108.05036 • Published • 3
-
TinyLLaVA: A Framework of Small-scale Large Multimodal Models
Paper • 2402.14289 • Published • 20 -
ImageBind: One Embedding Space To Bind Them All
Paper • 2305.05665 • Published • 6 -
DocLLM: A layout-aware generative language model for multimodal document understanding
Paper • 2401.00908 • Published • 191 -
Multimodal Contrastive Learning with LIMoE: the Language-Image Mixture of Experts
Paper • 2206.02770 • Published • 4
-
DocLLM: A layout-aware generative language model for multimodal document understanding
Paper • 2401.00908 • Published • 191 -
DeBERTa: Decoding-enhanced BERT with Disentangled Attention
Paper • 2006.03654 • Published • 3 -
DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing
Paper • 2111.09543 • Published • 3
-
FaceChain-SuDe: Building Derived Class to Inherit Category Attributes for One-shot Subject-Driven Generation
Paper • 2403.06775 • Published • 4 -
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Paper • 2010.11929 • Published • 15 -
Data Incubation -- Synthesizing Missing Data for Handwriting Recognition
Paper • 2110.07040 • Published • 2 -
A Mixture of Expert Approach for Low-Cost Customization of Deep Neural Networks
Paper • 1811.00056 • Published • 2
-
SpeechMoE: Scaling to Large Acoustic Models with Dynamic Routing Mixture of Experts
Paper • 2105.03036 • Published • 2 -
Building a great multi-lingual teacher with sparsely-gated mixture of experts for speech recognition
Paper • 2112.05820 • Published • 2 -
SpeechMoE2: Mixture-of-Experts Model with Improved Routing
Paper • 2111.11831 • Published • 2
-
QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models
Paper • 2310.16795 • Published • 27 -
Pareto-Optimal Quantized ResNet Is Mostly 4-bit
Paper • 2105.03536 • Published • 3 -
Decoding Compressed Trust: Scrutinizing the Trustworthiness of Efficient LLMs Under Compression
Paper • 2403.15447 • Published • 16
-
Data Incubation -- Synthesizing Missing Data for Handwriting Recognition
Paper • 2110.07040 • Published • 2 -
A Mixture of Expert Approach for Low-Cost Customization of Deep Neural Networks
Paper • 1811.00056 • Published • 2 -
Vulnerability Analysis of Transformer-based Optical Character Recognition to Adversarial Attacks
Paper • 2311.17128 • Published • 2 -
Data Generation for Post-OCR correction of Cyrillic handwriting
Paper • 2311.15896 • Published • 4
-
DocLLM: A layout-aware generative language model for multimodal document understanding
Paper • 2401.00908 • Published • 191 -
Visual Instruction Tuning
Paper • 2304.08485 • Published • 21 -
Glyph-ByT5: A Customized Text Encoder for Accurate Visual Text Rendering
Paper • 2403.09622 • Published • 17 -
Lumiere: A Space-Time Diffusion Model for Video Generation
Paper • 2401.12945 • Published • 86
-
Rescoring Sequence-to-Sequence Models for Text Line Recognition with CTC-Prefixes
Paper • 2110.05909 • Published • 2 -
Deep Residual Learning for Image Recognition
Paper • 1512.03385 • Published • 16 -
Wide Residual Networks
Paper • 1605.07146 • Published • 2 -
Comprehensive Survey of Model Compression and Speed up for Vision Transformers
Paper • 2404.10407 • Published • 1
-
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Paper • 1810.04805 • Published • 28 -
Transformers Can Achieve Length Generalization But Not Robustly
Paper • 2402.09371 • Published • 14 -
A Thorough Examination of Decoding Methods in the Era of LLMs
Paper • 2402.06925 • Published • 1 -
Byte Latent Transformer: Patches Scale Better Than Tokens
Paper • 2412.09871 • Published • 108
-
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
Paper • 1901.08746 • Published • 6 -
Pretraining-Based Natural Language Generation for Text Summarization
Paper • 1902.09243 • Published • 2 -
RoBERTa: A Robustly Optimized BERT Pretraining Approach
Paper • 1907.11692 • Published • 10 -
DeBERTa: Decoding-enhanced BERT with Disentangled Attention
Paper • 2006.03654 • Published • 3
-
Pretraining-Based Natural Language Generation for Text Summarization
Paper • 1902.09243 • Published • 2 -
Learning to Reason and Memorize with Self-Notes
Paper • 2305.00833 • Published • 5 -
Text Generation with Diffusion Language Models: A Pre-training Approach with Continuous Paragraph Denoise
Paper • 2212.11685 • Published • 2 -
Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems
Paper • 2408.16293 • Published • 27
-
An Interdisciplinary Comparison of Sequence Modeling Methods for Next-Element Prediction
Paper • 1811.00062 • Published • 2 -
mT5: A massively multilingual pre-trained text-to-text transformer
Paper • 2010.11934 • Published • 4 -
Bootstrap Your Own Skills: Learning to Solve New Tasks with Large Language Model Guidance
Paper • 2310.10021 • Published • 2 -
Gemma: Open Models Based on Gemini Research and Technology
Paper • 2403.08295 • Published • 50
-
Enhancing Document Information Analysis with Multi-Task Pre-training: A Robust Approach for Information Extraction in Visually-Rich Documents
Paper • 2310.16527 • Published • 2 -
CoDA: Collaborative Novel Box Discovery and Cross-modal Alignment for Open-vocabulary 3D Object Detection
Paper • 2310.02960 • Published • 2 -
MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training
Paper • 2403.09611 • Published • 129 -
Veagle: Advancements in Multimodal Representation Learning
Paper • 2403.08773 • Published • 10
-
Mixture-of-Supernets: Improving Weight-Sharing Supernet Training with Architecture-Routed Mixture-of-Experts
Paper • 2306.04845 • Published • 4 -
Balanced Mixture of SuperNets for Learning the CNN Pooling Architecture
Paper • 2306.11982 • Published • 2 -
AlphaNet: Improved Training of Supernets with Alpha-Divergence
Paper • 2102.07954 • Published • 2
-
Motion Mamba: Efficient and Long Sequence Motion Generation with Hierarchical and Bidirectional Selective SSM
Paper • 2403.07487 • Published • 16 -
LocalMamba: Visual State Space Model with Windowed Selective Scan
Paper • 2403.09338 • Published • 8 -
Cobra: Extending Mamba to Multi-Modal Large Language Model for Efficient Inference
Paper • 2403.14520 • Published • 35 -
SiMBA: Simplified Mamba-Based Architecture for Vision and Multivariate Time series
Paper • 2403.15360 • Published • 13
Mixtures of experts for text, image and speech
-
Chronos: Learning the Language of Time Series
Paper • 2403.07815 • Published • 50 -
Large Language Models as Optimizers
Paper • 2309.03409 • Published • 79 -
Pattern Discovery in Time Series with Byte Pair Encoding
Paper • 2106.00614 • Published • 2 -
MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection
Paper • 2403.19888 • Published • 12
-
AttentiveNAS: Improving Neural Architecture Search via Attentive Sampling
Paper • 2011.09011 • Published • 2 -
HAT: Hardware-Aware Transformers for Efficient Natural Language Processing
Paper • 2005.14187 • Published • 2 -
BigNAS: Scaling Up Neural Architecture Search with Big Single-Stage Models
Paper • 2003.11142 • Published • 2 -
Efficient Architecture Search by Network Transformation
Paper • 1707.04873 • Published • 2
-
U-Net: Convolutional Networks for Biomedical Image Segmentation
Paper • 1505.04597 • Published • 18 -
Image Segmentation using U-Net Architecture for Powder X-ray Diffraction Images
Paper • 2310.16186 • Published • 2 -
H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes
Paper • 1709.07330 • Published • 2 -
Deep LOGISMOS: Deep Learning Graph-based 3D Segmentation of Pancreatic Tumors on CT scans
Paper • 1801.08599 • Published • 2
-
H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes
Paper • 1709.07330 • Published • 2 -
Deep LOGISMOS: Deep Learning Graph-based 3D Segmentation of Pancreatic Tumors on CT scans
Paper • 1801.08599 • Published • 2 -
Hierarchical multi-class segmentation of glioma images using networks with multi-level activation function
Paper • 1810.09488 • Published • 2 -
Cross-modality (CT-MRI) prior augmented deep learning for robust lung tumor segmentation from small MR datasets
Paper • 1901.11369 • Published • 2
-
Enforcing Morphological Information in Fully Convolutional Networks to Improve Cell Instance Segmentation in Fluorescence Microscopy Images
Paper • 2106.05843 • Published • 2 -
Semi-Supervised Semantic Segmentation using Redesigned Self-Training for White Blood Cells
Paper • 2401.07278 • Published • 2
-
CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows
Paper • 2107.00652 • Published • 2 -
Cross-Shaped Windows Transformer with Self-supervised Pretraining for Clinically Significant Prostate Cancer Detection in Bi-parametric MRI
Paper • 2305.00385 • Published • 2 -
2nd Place Solution to Google Landmark Recognition Competition 2021
Paper • 2110.02638 • Published • 2 -
BOAT: Bilateral Local Attention Vision Transformer
Paper • 2201.13027 • Published • 2
-
MAFormer: A Transformer Network with Multi-scale Attention Fusion for Visual Recognition
Paper • 2209.01620 • Published • 2 -
Using Multi-scale SwinTransformer-HTC with Data augmentation in CoNIC Challenge
Paper • 2202.13588 • Published • 2 -
GasHis-Transformer: A Multi-scale Visual Transformer Approach for Gastric Histopathological Image Detection
Paper • 2104.14528 • Published • 2
-
Bootstrap Your Own Skills: Learning to Solve New Tasks with Large Language Model Guidance
Paper • 2310.10021 • Published • 2 -
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
Discovering Adaptable Symbolic Algorithms from Scratch
Paper • 2307.16890 • Published • 8 -
DragAPart: Learning a Part-Level Motion Prior for Articulated Objects
Paper • 2403.15382 • Published • 11
-
Feature-Guided Black-Box Safety Testing of Deep Neural Networks
Paper • 1710.07859 • Published • 2 -
Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks
Paper • 2309.17410 • Published • 4 -
Intriguing Properties of Adversarial Examples
Paper • 1711.02846 • Published • 2
-
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
JoMA: Demystifying Multilayer Transformers via JOint Dynamics of MLP and Attention
Paper • 2310.00535 • Published • 2 -
Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla
Paper • 2307.09458 • Published • 12 -
The Impact of Depth and Width on Transformer Language Model Generalization
Paper • 2310.19956 • Published • 10
-
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data
Paper • 2304.08247 • Published • 2 -
S-LoRA: Serving Thousands of Concurrent LoRA Adapters
Paper • 2311.03285 • Published • 30 -
WavLLM: Towards Robust and Adaptive Speech Large Language Model
Paper • 2404.00656 • Published • 11
-
UniAudio: An Audio Foundation Model Toward Universal Audio Generation
Paper • 2310.00704 • Published • 20 -
Structural Similarities Between Language Models and Neural Response Measurements
Paper • 2306.01930 • Published • 2 -
Streaming Transformer ASR with Blockwise Synchronous Beam Search
Paper • 2006.14941 • Published • 2 -
NU-GAN: High resolution neural upsampling with GAN
Paper • 2010.11362 • Published • 2
-
JoMA: Demystifying Multilayer Transformers via JOint Dynamics of MLP and Attention
Paper • 2310.00535 • Published • 2 -
Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small
Paper • 2211.00593 • Published • 2 -
Rethinking Interpretability in the Era of Large Language Models
Paper • 2402.01761 • Published • 23 -
Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla
Paper • 2307.09458 • Published • 12
-
A Biomedical Entity Extraction Pipeline for Oncology Health Records in Portuguese
Paper • 2304.08999 • Published • 3 -
CulturaX: A Cleaned, Enormous, and Multilingual Dataset for Large Language Models in 167 Languages
Paper • 2309.09400 • Published • 87 -
Robust Open-Vocabulary Translation from Visual Text Representations
Paper • 2104.08211 • Published • 1 -
Poro 34B and the Blessing of Multilinguality
Paper • 2404.01856 • Published • 15
-
LocalMamba: Visual State Space Model with Windowed Selective Scan
Paper • 2403.09338 • Published • 8 -
SiMBA: Simplified Mamba-Based Architecture for Vision and Multivariate Time series
Paper • 2403.15360 • Published • 13 -
MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection
Paper • 2403.19888 • Published • 12
-
Video Mamba Suite: State Space Model as a Versatile Alternative for Video Understanding
Paper • 2403.09626 • Published • 15 -
VideoAgent: Long-form Video Understanding with Large Language Model as Agent
Paper • 2403.10517 • Published • 37 -
VSTAR: Generative Temporal Nursing for Longer Dynamic Video Synthesis
Paper • 2403.13501 • Published • 9 -
LITA: Language Instructed Temporal-Localization Assistant
Paper • 2403.19046 • Published • 19
-
StreamMultiDiffusion: Real-Time Interactive Generation with Region-Based Semantic Control
Paper • 2403.09055 • Published • 26 -
GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models
Paper • 2112.10741 • Published • 4 -
Lightweight Image Inpainting by Stripe Window Transformer with Joint Attention to CNN
Paper • 2301.00553 • Published • 3 -
ObjectDrop: Bootstrapping Counterfactuals for Photorealistic Object Removal and Insertion
Paper • 2403.18818 • Published • 28
-
StreamMultiDiffusion: Real-Time Interactive Generation with Region-Based Semantic Control
Paper • 2403.09055 • Published • 26 -
ReNoise: Real Image Inversion Through Iterative Noising
Paper • 2403.14602 • Published • 21 -
EdgeFusion: On-Device Text-to-Image Generation
Paper • 2404.11925 • Published • 23
-
Contrastive Decoding Improves Reasoning in Large Language Models
Paper • 2309.09117 • Published • 40 -
Chain-of-Thought Reasoning Without Prompting
Paper • 2402.10200 • Published • 110 -
MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
Paper • 2403.14624 • Published • 53 -
Chain of Thought Empowers Transformers to Solve Inherently Serial Problems
Paper • 2402.12875 • Published • 13
-
Table-GPT: Table-tuned GPT for Diverse Table Tasks
Paper • 2310.09263 • Published • 40 -
approximatelabs/tablib-v1-full
Viewer • Updated • 10.4B • 36 • 67 -
approximatelabs/tablib-v1-sample
Viewer • Updated • 44.9k • 109 • 14 -
TabLib: A Dataset of 627M Tables with Context
Paper • 2310.07875 • Published • 8
-
MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries
Paper • 2401.15391 • Published • 6 -
Superposition Prompting: Improving and Accelerating Retrieval-Augmented Generation
Paper • 2404.06910 • Published • 3 -
Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models
Paper • 2404.18796 • Published • 71
-
CodeBERT: A Pre-Trained Model for Programming and Natural Languages
Paper • 2002.08155 • Published • 2 -
Text Generation with Diffusion Language Models: A Pre-training Approach with Continuous Paragraph Denoise
Paper • 2212.11685 • Published • 2 -
ReNoise: Real Image Inversion Through Iterative Noising
Paper • 2403.14602 • Published • 21 -
ByT5: Towards a token-free future with pre-trained byte-to-byte models
Paper • 2105.13626 • Published • 5
-
Neural networks behave as hash encoders: An empirical study
Paper • 2101.05490 • Published • 2 -
A Multiscale Visualization of Attention in the Transformer Model
Paper • 1906.05714 • Published • 2 -
BERT Rediscovers the Classical NLP Pipeline
Paper • 1905.05950 • Published • 3 -
Using Explainable AI and Transfer Learning to understand and predict the maintenance of Atlantic blocking with limited observational data
Paper • 2404.08613 • Published • 1
-
DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining
Paper • 2305.10429 • Published • 5 -
LLM2LLM: Boosting LLMs with Novel Iterative Data Enhancement
Paper • 2403.15042 • Published • 27 -
Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models
Paper • 2405.01535 • Published • 124 -
Discovering Preference Optimization Algorithms with and for Large Language Models
Paper • 2406.08414 • Published • 16
-
Adafactor: Adaptive Learning Rates with Sublinear Memory Cost
Paper • 1804.04235 • Published • 2 -
DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining
Paper • 2305.10429 • Published • 5 -
GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
Paper • 2305.13245 • Published • 6
-
DocVQA: A Dataset for VQA on Document Images
Paper • 2007.00398 • Published • 2 -
GQA: A New Dataset for Real-World Visual Reasoning and Compositional Question Answering
Paper • 1902.09506 • Published • 2 -
google/wit
Viewer • Updated • 2.66M • 468 • 68 -
Lin-Chen/MMStar
Viewer • Updated • 1.5k • 31k • 51
-
Adding Conditional Control to Text-to-Image Diffusion Models
Paper • 2302.05543 • Published • 58 -
LightIt: Illumination Modeling and Control for Diffusion Models
Paper • 2403.10615 • Published • 18 -
SDXS: Real-Time One-Step Latent Diffusion Models with Image Conditions
Paper • 2403.16627 • Published • 22 -
DreamPolisher: Towards High-Quality Text-to-3D Generation via Geometric Diffusion
Paper • 2403.17237 • Published • 11
-
PERL: Parameter Efficient Reinforcement Learning from Human Feedback
Paper • 2403.10704 • Published • 60 -
WARM: On the Benefits of Weight Averaged Reward Models
Paper • 2401.12187 • Published • 19 -
RewardBench: Evaluating Reward Models for Language Modeling
Paper • 2403.13787 • Published • 22 -
DreamReward: Text-to-3D Generation with Human Preference
Paper • 2403.14613 • Published • 37
-
Vid2Robot: End-to-end Video-conditioned Policy Learning with Cross-Attention Transformers
Paper • 2403.12943 • Published • 15 -
Masked Audio Generation using a Single Non-Autoregressive Transformer
Paper • 2401.04577 • Published • 44 -
Cross-Attention Makes Inference Cumbersome in Text-to-Image Diffusion Models
Paper • 2404.02747 • Published • 13 -
InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation
Paper • 2404.02733 • Published • 22
-
LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression
Paper • 2403.12968 • Published • 26 -
Improving Text-to-Image Consistency via Automatic Prompt Optimization
Paper • 2403.17804 • Published • 19 -
The Unreasonable Effectiveness of Eccentric Automatic Prompts
Paper • 2402.10949 • Published • 5 -
Orca: Progressive Learning from Complex Explanation Traces of GPT-4
Paper • 2306.02707 • Published • 51
-
TexDreamer: Towards Zero-Shot High-Fidelity 3D Human Texture Generation
Paper • 2403.12906 • Published • 7 -
GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generation
Paper • 2403.14621 • Published • 16 -
LATTE3D: Large-scale Amortized Text-To-Enhanced3D Synthesis
Paper • 2403.15385 • Published • 8 -
LLaMA-Mesh: Unifying 3D Mesh Generation with Language Models
Paper • 2411.09595 • Published • 77
Converted the Elephants Never Forget paper to audio with Bark: https://drive.google.com/file/d/13IlbhKh71vxLpdYJ6mkIiiJZOUsf7XFv/view?usp=sharing
-
End-to-End Object Detection with Transformers
Paper • 2005.12872 • Published • 7 -
Elephants Never Forget: Memorization and Learning of Tabular Data in Large Language Models
Paper • 2404.06209 • Published • 5 -
TabReD: A Benchmark of Tabular Machine Learning in-the-Wild
Paper • 2406.19380 • Published • 49 -
SpreadsheetLLM: Encoding Spreadsheets for Large Language Models
Paper • 2407.09025 • Published • 140
-
Evolutionary Optimization of Model Merging Recipes
Paper • 2403.13187 • Published • 58 -
Model Stock: All we need is just a few fine-tuned models
Paper • 2403.19522 • Published • 15 -
Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models
Paper • 2405.01535 • Published • 124
-
MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
Paper • 2403.14624 • Published • 53 -
We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?
Paper • 2407.01284 • Published • 81 -
MAVIS: Mathematical Visual Instruction Tuning
Paper • 2407.08739 • Published • 32
-
Recourse for reclamation: Chatting with generative language models
Paper • 2403.14467 • Published • 8 -
Decoding Compressed Trust: Scrutinizing the Trustworthiness of Efficient LLMs Under Compression
Paper • 2403.15447 • Published • 16 -
Introducing v0.5 of the AI Safety Benchmark from MLCommons
Paper • 2404.12241 • Published • 13
-
Lexicon-Level Contrastive Visual-Grounding Improves Language Modeling
Paper • 2403.14551 • Published • 2 -
Adapting LLaMA Decoder to Vision Transformer
Paper • 2404.06773 • Published • 19 -
Toward a Better Understanding of Fourier Neural Operators: Analysis and Improvement from a Spectral Perspective
Paper • 2404.07200 • Published • 2 -
An inclusive review on deep learning techniques and their scope in handwriting recognition
Paper • 2404.08011 • Published • 1
-
LLM-Assisted Content Analysis: Using Large Language Models to Support Deductive Coding
Paper • 2306.14924 • Published • 2 -
When LLMs are Unfit Use FastFit: Fast and Effective Text Classification with Many Classes
Paper • 2404.12365 • Published • 2 -
In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering
Paper • 2311.06668 • Published • 5 -
Wave Network: An Ultra-Small Language Model
Paper • 2411.02674 • Published • 3
-
A Multimodal Approach to Device-Directed Speech Detection with Large Language Models
Paper • 2403.14438 • Published • 2 -
AniPortrait: Audio-Driven Synthesis of Photorealistic Portrait Animation
Paper • 2403.17694 • Published • 12 -
FlashSpeech: Efficient Zero-Shot Speech Synthesis
Paper • 2404.14700 • Published • 32 -
SONAR: Sentence-Level Multimodal and Language-Agnostic Representations
Paper • 2308.11466 • Published • 1
-
Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models
Paper • 2311.00871 • Published • 3 -
Can large language models explore in-context?
Paper • 2403.15371 • Published • 33 -
Data Distributional Properties Drive Emergent In-Context Learning in Transformers
Paper • 2205.05055 • Published • 2 -
Long-context LLMs Struggle with Long In-context Learning
Paper • 2404.02060 • Published • 37
Refer to additional papers: https://link.springer.com/article/10.1007/s10994-014-5458-8 and https://link.springer.com/article/10.1007/BF00992696
-
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
Paper • 2305.18290 • Published • 66 -
ICDPO: Effectively Borrowing Alignment Capability of Others via In-context Direct Preference Optimization
Paper • 2402.09320 • Published • 6 -
sDPO: Don't Use Your Data All at Once
Paper • 2403.19270 • Published • 41 -
Dueling RL: Reinforcement Learning with Trajectory Preferences
Paper • 2111.04850 • Published • 2
-
CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing
Paper • 2305.11738 • Published • 9 -
Shepherd: A Critic for Language Model Generation
Paper • 2308.04592 • Published • 33 -
CriticBench: Benchmarking LLMs for Critique-Correct Reasoning
Paper • 2402.14809 • Published • 3 -
DRLC: Reinforcement Learning with Dense Rewards from LLM Critic
Paper • 2401.07382 • Published • 2
-
CriticBench: Benchmarking LLMs for Critique-Correct Reasoning
Paper • 2402.14809 • Published • 3 -
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
TACT: Advancing Complex Aggregative Reasoning with Information Extraction Tools
Paper • 2406.03618 • Published • 2
-
ReGAL: Refactoring Programs to Discover Generalizable Abstractions
Paper • 2401.16467 • Published • 10 -
Language Models as Compilers: Simulating Pseudocode Execution Improves Algorithmic Reasoning in Language Models
Paper • 2404.02575 • Published • 50 -
How Far Can We Go with Practical Function-Level Program Repair?
Paper • 2404.12833 • Published • 7
-
V-STaR: Training Verifiers for Self-Taught Reasoners
Paper • 2402.06457 • Published • 9 -
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
McEval: Massively Multilingual Code Evaluation
Paper • 2406.07436 • Published • 41 -
Is Programming by Example solved by LLMs?
Paper • 2406.08316 • Published • 13
-
V-STaR: Training Verifiers for Self-Taught Reasoners
Paper • 2402.06457 • Published • 9 -
Self-Consistency Preference Optimization
Paper • 2411.04109 • Published • 19 -
Marco-o1: Towards Open Reasoning Models for Open-Ended Solutions
Paper • 2411.14405 • Published • 61 -
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Paper • 2402.03300 • Published • 145
-
ThemeStation: Generating Theme-Aware 3D Assets from Few Exemplars
Paper • 2403.15383 • Published • 15 -
LATTE3D: Large-scale Amortized Text-To-Enhanced3D Synthesis
Paper • 2403.15385 • Published • 8 -
MaPa: Text-driven Photorealistic Material Painting for 3D Shapes
Paper • 2404.17569 • Published • 13 -
PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning
Paper • 2404.16994 • Published • 37
-
FollowIR: Evaluating and Teaching Information Retrieval Models to Follow Instructions
Paper • 2403.15246 • Published • 11 -
Noise-Aware Training of Layout-Aware Language Models
Paper • 2404.00488 • Published • 10 -
Text Role Classification in Scientific Charts Using Multimodal Transformers
Paper • 2402.14579 • Published • 1
-
Compiler generated feedback for Large Language Models
Paper • 2403.14714 • Published • 7 -
Language Models as Compilers: Simulating Pseudocode Execution Improves Algorithmic Reasoning in Language Models
Paper • 2404.02575 • Published • 50 -
Compiling C to Safe Rust, Formalized
Paper • 2412.15042 • Published • 1
-
EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters
Paper • 2402.04252 • Published • 30 -
MiniGPT4-Video: Advancing Multimodal LLMs for Video Understanding with Interleaved Visual-Textual Tokens
Paper • 2404.03413 • Published • 27 -
openai/clip-vit-large-patch14-336
Zero-Shot Image Classification • Updated • 9.33M • 307 -
openai/clip-vit-base-patch32
Zero-Shot Image Classification • Updated • 21.5M • 933
-
Insightful analysis of historical sources at scales beyond human capabilities using unsupervised Machine Learning and XAI
Paper • 2310.09091 • Published • 2 -
Evolution and Transformation of Scientific Knowledge over the Sphaera Corpus: A Network Study
Paper • 2004.00520 • Published • 2 -
NAF-DPM: A Nonlinear Activation-Free Diffusion Probabilistic Model for Document Enhancement
Paper • 2404.05669 • Published • 1
-
InternLM2 Technical Report
Paper • 2403.17297 • Published • 34 -
sDPO: Don't Use Your Data All at Once
Paper • 2403.19270 • Published • 41 -
Learn Your Reference Model for Real Good Alignment
Paper • 2404.09656 • Published • 90 -
OpenBezoar: Small, Cost-Effective and Open Models Trained on Mixes of Instruction Data
Paper • 2404.12195 • Published • 12
-
Lumiere: A Space-Time Diffusion Model for Video Generation
Paper • 2401.12945 • Published • 86 -
Long-form factuality in large language models
Paper • 2403.18802 • Published • 26 -
ObjectDrop: Bootstrapping Counterfactuals for Photorealistic Object Removal and Insertion
Paper • 2403.18818 • Published • 28 -
TC4D: Trajectory-Conditioned Text-to-4D Generation
Paper • 2403.17920 • Published • 18
-
The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
Paper • 1801.03924 • Published • 2 -
LLM2LLM: Boosting LLMs with Novel Iterative Data Enhancement
Paper • 2403.15042 • Published • 27 -
Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions
Paper • 1712.05884 • Published • 3 -
Efficient Memory Management for Large Language Model Serving with PagedAttention
Paper • 2309.06180 • Published • 54
-
Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction
Paper • 2403.18795 • Published • 20 -
EgoLifter: Open-world 3D Segmentation for Egocentric Perception
Paper • 2403.18118 • Published • 12 -
GaussianCube: Structuring Gaussian Splatting using Optimal Transport for 3D Generative Modeling
Paper • 2403.19655 • Published • 19 -
Snap-it, Tap-it, Splat-it: Tactile-Informed 3D Gaussian Splatting for Reconstructing Challenging Surfaces
Paper • 2403.20275 • Published • 10
-
Towards a World-English Language Model for On-Device Virtual Assistants
Paper • 2403.18783 • Published • 6 -
MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training
Paper • 2403.09611 • Published • 129 -
ReALM: Reference Resolution As Language Modeling
Paper • 2403.20329 • Published • 22 -
Ferret-UI: Grounded Mobile UI Understanding with Multimodal LLMs
Paper • 2404.05719 • Published • 83
-
Can large language models explore in-context?
Paper • 2403.15371 • Published • 33 -
GaussianCube: Structuring Gaussian Splatting using Optimal Transport for 3D Generative Modeling
Paper • 2403.19655 • Published • 19 -
WavLLM: Towards Robust and Adaptive Speech Large Language Model
Paper • 2404.00656 • Published • 11 -
Enabling Memory Safety of C Programs using LLMs
Paper • 2404.01096 • Published • 1
-
Generating Synthetic Computed Tomography for Radiotherapy: SynthRAD2023 Challenge Report
Paper • 2403.08447 • Published • 2 -
Deformable MRI Sequence Registration for AI-based Prostate Cancer Diagnosis
Paper • 2404.09666 • Published • 1 -
Surgical SAM 2: Real-time Segment Anything in Surgical Video by Efficient Frame Pruning
Paper • 2408.07931 • Published • 22
-
Mesh2NeRF: Direct Mesh Supervision for Neural Radiance Field Representation and Generation
Paper • 2403.19319 • Published • 14 -
Getting it Right: Improving Spatial Consistency in Text-to-Image Models
Paper • 2404.01197 • Published • 31 -
LLaVA-Gemma: Accelerating Multimodal Foundation Models with a Compact Language Model
Paper • 2404.01331 • Published • 27 -
LVLM-Intrepret: An Interpretability Tool for Large Vision-Language Models
Paper • 2404.03118 • Published • 25
-
Dueling RL: Reinforcement Learning with Trajectory Preferences
Paper • 2111.04850 • Published • 2 -
Learning Trajectory Preferences for Manipulators via Iterative Improvement
Paper • 1306.6294 • Published • 3 -
Deep reinforcement learning from human preferences
Paper • 1706.03741 • Published • 4 -
Learning Dynamic Robot-to-Human Object Handover from Human Feedback
Paper • 1603.06390 • Published • 2
-
Fine-Tuning Language Models from Human Preferences
Paper • 1909.08593 • Published • 4 -
Transforming and Combining Rewards for Aligning Large Language Models
Paper • 2402.00742 • Published • 12 -
Leverage the Average: an Analysis of KL Regularization in RL
Paper • 2003.14089 • Published • 2 -
Direct Preference Optimization of Video Large Multimodal Models from Language Model Reward
Paper • 2404.01258 • Published • 12
-
Transforming and Combining Rewards for Aligning Large Language Models
Paper • 2402.00742 • Published • 12 -
UltraFeedback: Boosting Language Models with High-quality Feedback
Paper • 2310.01377 • Published • 6 -
Learn Your Reference Model for Real Good Alignment
Paper • 2404.09656 • Published • 90 -
Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models
Paper • 2405.01535 • Published • 124
Figure 5: The probability mass assigned to partial human sentences. Flat distributions lead to many
moderately probable tokens, while peaked distribut
-
The Curious Case of Neural Text Degeneration
Paper • 1904.09751 • Published • 3 -
PIQA: Reasoning about Physical Commonsense in Natural Language
Paper • 1911.11641 • Published • 5 -
SocialIQA: Commonsense Reasoning about Social Interactions
Paper • 1904.09728 • Published • 4 -
HellaSwag: Can a Machine Really Finish Your Sentence?
Paper • 1905.07830 • Published • 8
-
Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions
Paper • 1712.05884 • Published • 3 -
VoiceCraft: Zero-Shot Speech Editing and Text-to-Speech in the Wild
Paper • 2403.16973 • Published • 3 -
High Fidelity Neural Audio Compression
Paper • 2210.13438 • Published • 4 -
RALL-E: Robust Codec Language Modeling with Chain-of-Thought Prompting for Text-to-Speech Synthesis
Paper • 2404.03204 • Published • 9
-
keeeeenw/MicroLlama
Text Generation • 0.3B • Updated • 270 • 53 -
TinyLlama/TinyLlama-1.1B-intermediate-step-240k-503b
Text Generation • 1B • Updated • 1.49k • 20 -
TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
Text Generation • Updated • 105k • 188 -
TinyLlama/TinyLlama-1.1B-Chat-v1.0
Text Generation • 1B • Updated • 2.87M • • 1.58k
-
Gecko: Versatile Text Embeddings Distilled from Large Language Models
Paper • 2403.20327 • Published • 47 -
Round and Round We Go! What makes Rotary Positional Encodings useful?
Paper • 2410.06205 • Published • 2 -
Byte Latent Transformer: Patches Scale Better Than Tokens
Paper • 2412.09871 • Published • 108 -
MrT5: Dynamic Token Merging for Efficient Byte-level Language Models
Paper • 2410.20771 • Published • 3
Gradients flow differently for memorized and non-memorized during decoding
-
DiJiang: Efficient Large Language Models through Compact Kernelization
Paper • 2403.19928 • Published • 12 -
MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
Paper • 2309.12284 • Published • 19 -
TextHawk: Exploring Efficient Fine-Grained Perception of Multimodal Large Language Models
Paper • 2404.09204 • Published • 11 -
SAGS: Structure-Aware 3D Gaussian Splatting
Paper • 2404.19149 • Published • 14
-
The FinBen: An Holistic Financial Benchmark for Large Language Models
Paper • 2402.12659 • Published • 24 -
Long-context LLMs Struggle with Long In-context Learning
Paper • 2404.02060 • Published • 37 -
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding
Paper • 1804.07461 • Published • 4
-
DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
Paper • 1606.06160 • Published • 1 -
The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits
Paper • 2402.17764 • Published • 629 -
mobiuslabsgmbh/Llama-2-7b-chat-hf_1bitgs8_hqq
Text Generation • Updated • 25 • 74
-
Condition-Aware Neural Network for Controlled Image Generation
Paper • 2404.01143 • Published • 13 -
FlexiDreamer: Single Image-to-3D Generation with FlexiCubes
Paper • 2404.00987 • Published • 23 -
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline
Paper • 2404.02893 • Published • 21
-
WavLLM: Towards Robust and Adaptive Speech Large Language Model
Paper • 2404.00656 • Published • 11 -
Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization
Paper • 2404.09956 • Published • 11 -
Long-form music generation with latent diffusion
Paper • 2404.10301 • Published • 27 -
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations
Paper • 2006.11477 • Published • 9
-
Noise-Aware Training of Layout-Aware Language Models
Paper • 2404.00488 • Published • 10 -
FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extraction
Paper • 2305.02549 • Published • 7 -
FormNet: Structural Encoding beyond Sequential Modeling in Form Document Information Extraction
Paper • 2203.08411 • Published • 1 -
ETC: Encoding Long and Structured Inputs in Transformers
Paper • 2004.08483 • Published • 1
-
The state-of-the-art in Cardiac MRI Reconstruction: Results of the CMRxRecon Challenge in MICCAI 2023
Paper • 2404.01082 • Published • 1 -
Realism in Action: Anomaly-Aware Diagnosis of Brain Tumors from Medical Images Using YOLOv8 and DeiT
Paper • 2401.03302 • Published • 1 -
Brain2Music: Reconstructing Music from Human Brain Activity
Paper • 2307.11078 • Published • 42
-
FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extraction
Paper • 2305.02549 • Published • 7 -
FormNet: Structural Encoding beyond Sequential Modeling in Form Document Information Extraction
Paper • 2203.08411 • Published • 1 -
More efficient manual review of automatically transcribed tabular data
Paper • 2306.16126 • Published • 1 -
CascadeTabNet: An approach for end to end table detection and structure recognition from image-based documents
Paper • 2004.12629 • Published • 3
https://github.com/learn2phoenix/CSD
-
CameraCtrl: Enabling Camera Control for Text-to-Video Generation
Paper • 2404.02101 • Published • 24 -
Adapting LLaMA Decoder to Vision Transformer
Paper • 2404.06773 • Published • 19 -
Interactive3D: Create What You Want by Interactive 3D Generation
Paper • 2404.16510 • Published • 21 -
Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
Paper • 2406.07394 • Published • 29
-
Long-context LLMs Struggle with Long In-context Learning
Paper • 2404.02060 • Published • 37 -
Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks
Paper • 2211.12588 • Published • 3 -
StructLM: Towards Building Generalist Models for Structured Knowledge Grounding
Paper • 2402.16671 • Published • 27 -
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model
Paper • 2404.04167 • Published • 13
https://github.com/MoritzLaurer/zeroshot-classifier/tree/main
-
LLM-ABR: Designing Adaptive Bitrate Algorithms via Large Language Models
Paper • 2404.01617 • Published • 8 -
Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction
Paper • 2404.02905 • Published • 74 -
Learning From Mistakes Makes LLM Better Reasoner
Paper • 2310.20689 • Published • 29 -
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model
Paper • 2404.04167 • Published • 13
-
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
PointInfinity: Resolution-Invariant Point Diffusion Models
Paper • 2404.03566 • Published • 16 -
MonoPatchNeRF: Improving Neural Radiance Fields with Patch-based Monocular Guidance
Paper • 2404.08252 • Published • 6 -
SnapKV: LLM Knows What You are Looking for Before Generation
Paper • 2404.14469 • Published • 27
-
LLaVA-Gemma: Accelerating Multimodal Foundation Models with a Compact Language Model
Paper • 2404.01331 • Published • 27 -
LVLM-Intrepret: An Interpretability Tool for Large Vision-Language Models
Paper • 2404.03118 • Published • 25 -
DesignQA: A Multimodal Benchmark for Evaluating Large Language Models' Understanding of Engineering Documentation
Paper • 2404.07917 • Published • 3 -
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32
-
LLaVA-Gemma: Accelerating Multimodal Foundation Models with a Compact Language Model
Paper • 2404.01331 • Published • 27 -
Data curation via joint example selection further accelerates multimodal learning
Paper • 2406.17711 • Published • 3 -
Unveiling Encoder-Free Vision-Language Models
Paper • 2406.11832 • Published • 55
-
LLaVA-Gemma: Accelerating Multimodal Foundation Models with a Compact Language Model
Paper • 2404.01331 • Published • 27 -
OmniFusion Technical Report
Paper • 2404.06212 • Published • 77 -
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15 -
WildGaussians: 3D Gaussian Splatting in the Wild
Paper • 2407.08447 • Published • 9
For HyperClova X they split 90% at 4096 and 10% at 32k context length during pt
Batches are grouped by similar token length to help optimize gpu/hardware. Mini batch lengths are different but the max number of tokens is the same.
-
On the Scalability of Diffusion-based Text-to-Image Generation
Paper • 2404.02883 • Published • 19 -
InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation
Paper • 2404.02733 • Published • 22 -
CoMat: Aligning Text-to-Image Diffusion Model with Image-to-Text Concept Matching
Paper • 2404.03653 • Published • 35 -
ControlNet++: Improving Conditional Controls with Efficient Consistency Feedback
Paper • 2404.07987 • Published • 48
-
Freditor: High-Fidelity and Transferable NeRF Editing by Frequency Decomposition
Paper • 2404.02514 • Published • 11 -
BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer
Paper • 1904.06690 • Published • 1 -
Contrastive Chain-of-Thought Prompting
Paper • 2311.09277 • Published • 35 -
LLM-R2: A Large Language Model Enhanced Rule-based Rewrite System for Boosting Query Efficiency
Paper • 2404.12872 • Published • 11
-
Event Camera Demosaicing via Swin Transformer and Pixel-focus Loss
Paper • 2404.02731 • Published • 1 -
MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
Paper • 2309.12284 • Published • 19 -
RALL-E: Robust Codec Language Modeling with Chain-of-Thought Prompting for Text-to-Speech Synthesis
Paper • 2404.03204 • Published • 9 -
Adapting LLaMA Decoder to Vision Transformer
Paper • 2404.06773 • Published • 19
-
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing
Paper • 2404.12253 • Published • 55 -
SEED-Bench-2-Plus: Benchmarking Multimodal Large Language Models with Text-Rich Visual Comprehension
Paper • 2404.16790 • Published • 10 -
A Thorough Examination of Decoding Methods in the Era of LLMs
Paper • 2402.06925 • Published • 1
Table 5: Transformers struggle to learn Arithmetic Coding. In the sequence-to-sequence setting,
a model that learns AC compression/decompression shoul
Emergence with scale is unlikely Given the recent findings of [55], we anticipate that continuing
to scale models beyond 2 billion parameters is unlik
In this paper, we propose a strong alternative to PEFTs, LoReFT. LoReFT achieves strong per-
formance across benchmarks from four domains while being
-
SocialIQA: Commonsense Reasoning about Social Interactions
Paper • 1904.09728 • Published • 4 -
PIQA: Reasoning about Physical Commonsense in Natural Language
Paper • 1911.11641 • Published • 5 -
BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions
Paper • 1905.10044 • Published • 2 -
HellaSwag: Can a Machine Really Finish Your Sentence?
Paper • 1905.07830 • Published • 8
-
AQuA: A Benchmarking Tool for Label Quality Assessment
Paper • 2306.09467 • Published • 1 -
OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
Paper • 2404.07972 • Published • 52 -
BLINK: Multimodal Large Language Models Can See but Not Perceive
Paper • 2404.12390 • Published • 26 -
Vision language models are blind
Paper • 2407.06581 • Published • 84
https://github.com/google-deepmind/AQuA
-
RoBERTa: A Robustly Optimized BERT Pretraining Approach
Paper • 1907.11692 • Published • 10 -
Leveraging Pre-trained Checkpoints for Sequence Generation Tasks
Paper • 1907.12461 • Published • 1 -
Transformer Language Models without Positional Encodings Still Learn Positional Information
Paper • 2203.16634 • Published • 6 -
CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generation
Paper • 2102.04664 • Published • 2
-
MiniGPT4-Video: Advancing Multimodal LLMs for Video Understanding with Interleaved Visual-Textual Tokens
Paper • 2404.03413 • Published • 27 -
Scaling Data-Constrained Language Models
Paper • 2305.16264 • Published • 16 -
Emergence of Hidden Capabilities: Exploring Learning Dynamics in Concept Space
Paper • 2406.19370 • Published • 1
-
RALL-E: Robust Codec Language Modeling with Chain-of-Thought Prompting for Text-to-Speech Synthesis
Paper • 2404.03204 • Published • 9 -
Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models
Paper • 2311.07919 • Published • 9 -
FlashSpeech: Efficient Zero-Shot Speech Synthesis
Paper • 2404.14700 • Published • 32
-
Prompt-to-Prompt Image Editing with Cross Attention Control
Paper • 2208.01626 • Published • 3 -
BERT Rediscovers the Classical NLP Pipeline
Paper • 1905.05950 • Published • 3 -
A Multiscale Visualization of Attention in the Transformer Model
Paper • 1906.05714 • Published • 2 -
Analyzing Transformers in Embedding Space
Paper • 2209.02535 • Published • 3
-
DeViDe: Faceted medical knowledge for improved medical vision-language pre-training
Paper • 2404.03618 • Published • 2 -
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model
Paper • 2404.04167 • Published • 13 -
SpecInfer: Accelerating Generative LLM Serving with Speculative Inference and Token Tree Verification
Paper • 2305.09781 • Published • 4 -
McEval: Massively Multilingual Code Evaluation
Paper • 2406.07436 • Published • 41
Reward expressed as win-rates related to general preferences
-
Realism in Action: Anomaly-Aware Diagnosis of Brain Tumors from Medical Images Using YOLOv8 and DeiT
Paper • 2401.03302 • Published • 1 -
MLP Can Be A Good Transformer Learner
Paper • 2404.05657 • Published • 1 -
Detecting and recognizing characters in Greek papyri with YOLOv8, DeiT and SimCLR
Paper • 2401.12513 • Published • 1 -
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets
Paper • 2404.02900 • Published • 1
-
Kandinsky: an Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion
Paper • 2310.03502 • Published • 79 -
Transferable and Principled Efficiency for Open-Vocabulary Segmentation
Paper • 2404.07448 • Published • 12 -
RegionGPT: Towards Region Understanding Vision Language Model
Paper • 2403.02330 • Published • 2 -
GLIGEN: Open-Set Grounded Text-to-Image Generation
Paper • 2301.07093 • Published • 4
-
Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention
Paper • 2404.07143 • Published • 111 -
RULER: What's the Real Context Size of Your Long-Context Language Models?
Paper • 2404.06654 • Published • 41 -
An Evolved Universal Transformer Memory
Paper • 2410.13166 • Published • 6
-
DreamScene360: Unconstrained Text-to-3D Scene Generation with Panoramic Gaussian Splatting
Paper • 2404.06903 • Published • 21 -
CatLIP: CLIP-level Visual Recognition Accuracy with 2.7x Faster Pre-training on Web-scale Image-Text Data
Paper • 2404.15653 • Published • 29 -
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15 -
BlenderAlchemy: Editing 3D Graphics with Vision-Language Models
Paper • 2404.17672 • Published • 19
-
End-to-End Object Detection with Transformers
Paper • 2005.12872 • Published • 7 -
ConsistencyDet: Robust Object Detector with Denoising Paradigm of Consistency Model
Paper • 2404.07773 • Published • 1 -
Efficient Transformer Encoders for Mask2Former-style models
Paper • 2404.15244 • Published • 1 -
DETRs Beat YOLOs on Real-time Object Detection
Paper • 2304.08069 • Published • 16
-
ORPO: Monolithic Preference Optimization without Reference Model
Paper • 2403.07691 • Published • 72 -
ResearchAgent: Iterative Research Idea Generation over Scientific Literature with Large Language Models
Paper • 2404.07738 • Published • 2 -
Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models
Paper • 2405.01535 • Published • 124
-
JetMoE: Reaching Llama2 Performance with 0.1M Dollars
Paper • 2404.07413 • Published • 38 -
Allowing humans to interactively guide machines where to look does not always improve a human-AI team's classification accuracy
Paper • 2404.05238 • Published • 3 -
Cognitive Architectures for Language Agents
Paper • 2309.02427 • Published • 8 -
Latent Positional Information is in the Self-Attention Variance of Transformer Language Models Without Positional Embeddings
Paper • 2305.13571 • Published • 2
Generalized multi head using RoPE
Read by Bark: https://drive.google.com/file/d/1apmyvLMEQ97ObHKzQna9URFHF0Xg-EsO/view?usp=sharing
-
Rho-1: Not All Tokens Are What You Need
Paper • 2404.07965 • Published • 94 -
Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems
Paper • 2408.16293 • Published • 27 -
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Paper • 2402.03300 • Published • 145
-
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32 -
RegionGPT: Towards Region Understanding Vision Language Model
Paper • 2403.02330 • Published • 2 -
TextSquare: Scaling up Text-Centric Visual Instruction Tuning
Paper • 2404.12803 • Published • 30 -
Pegasus-v1 Technical Report
Paper • 2404.14687 • Published • 33
Where the model is tasked with identifying
the object in a region mentioned in a query. we utilize the validation split of the
LVIS dataset
Read by Bark: https://drive.google.com/file/d/1no3kjSmexQxlS-KjhRB0jB5hz72Yuhsb/view?usp=sharing
Read by Bark: https://drive.google.com/file/d/13IlbhKh71vxLpdYJ6mkIiiJZOUsf7XFv/view?usp=sharing
-
mlabonne/AlphaMonarch-7B
Text Generation • 7B • Updated • 13.4k • • 148 -
Qwen/QwQ-32B-Preview
Text Generation • 33B • Updated • 10.3k • • 1.74k -
deepseek-ai/DeepSeek-R1-Distill-Llama-8B
Text Generation • 8B • Updated • 867k • • 859 -
deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
Text Generation • 8B • Updated • 567k • • 831
-
Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization
Paper • 2404.09956 • Published • 11 -
Contrastive Chain-of-Thought Prompting
Paper • 2311.09277 • Published • 35 -
FlashSpeech: Efficient Zero-Shot Speech Synthesis
Paper • 2404.14700 • Published • 32 -
AsyncDiff: Parallelizing Diffusion Models by Asynchronous Denoising
Paper • 2406.06911 • Published • 12
We use an ensemble filtering strategy based on two different CLAP models: 630k-audioset-best and 630k-best
-
ResearchAgent: Iterative Research Idea Generation over Scientific Literature with Large Language Models
Paper • 2404.07738 • Published • 2 -
Ag2Manip: Learning Novel Manipulation Skills with Agent-Agnostic Visual and Action Representations
Paper • 2404.17521 • Published • 13 -
LEGENT: Open Platform for Embodied Agents
Paper • 2404.18243 • Published • 22 -
Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
Paper • 2406.07394 • Published • 29
Read by Bark: https://drive.google.com/file/d/1yCc6rr199rQHHNwKozHhqzz0Rr48z03A/view?usp=sharing duration is 1 hour and 11 min 47s
better loss function. The sigmoid loss operates solely on image-text pairs and does not require a global view of the pairwise similarities
Found 5566 memorized, repeated audio sequences
The embeddings of
images and video patches from raw frames x are processed by a spatial and then a causal transformer
(AR in time) to gen video tokens
Grad-CAM
We propose to perform reward optimization using a RM trained for a different language. Assuming model generation quality transfers cross-lingually
multilingual, pairwise human-rated chat transcripts.
For the SFT data, we use the human-preferred response in each pair to finetune the model
we recognize two memory bottlenecks: model weights and KV cache, and the latter gradually bottleneck(s) as context length increases
-
Dynamic Typography: Bringing Words to Life
Paper • 2404.11614 • Published • 46 -
Scene Coordinate Reconstruction: Posing of Image Collections via Incremental Learning of a Relocalizer
Paper • 2404.14351 • Published • 6 -
BlenderAlchemy: Editing 3D Graphics with Vision-Language Models
Paper • 2404.17672 • Published • 19 -
Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation
Paper • 2406.06525 • Published • 71
-
Length Generalization of Causal Transformers without Position Encoding
Paper • 2404.12224 • Published • 1 -
Transformer Language Models without Positional Encodings Still Learn Positional Information
Paper • 2203.16634 • Published • 6 -
Latent Positional Information is in the Self-Attention Variance of Transformer Language Models Without Positional Embeddings
Paper • 2305.13571 • Published • 2 -
The Impact of Positional Encoding on Length Generalization in Transformers
Paper • 2305.19466 • Published • 2
-
Latent Positional Information is in the Self-Attention Variance of Transformer Language Models Without Positional Embeddings
Paper • 2305.13571 • Published • 2 -
Transformers Can Represent n-gram Language Models
Paper • 2404.14994 • Published • 21 -
Are Sixteen Heads Really Better than One?
Paper • 1905.10650 • Published • 2 -
Reasoning in Large Language Models: A Geometric Perspective
Paper • 2407.02678 • Published • 1
-
Latent Positional Information is in the Self-Attention Variance of Transformer Language Models Without Positional Embeddings
Paper • 2305.13571 • Published • 2 -
BERTs are Generative In-Context Learners
Paper • 2406.04823 • Published • 1 -
Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference
Paper • 2412.13663 • Published • 163 -
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations
Paper • 2006.11477 • Published • 9
-
The Impact of Positional Encoding on Length Generalization in Transformers
Paper • 2305.19466 • Published • 2 -
Qwen2 Technical Report
Paper • 2407.10671 • Published • 172 -
Round and Round We Go! What makes Rotary Positional Encodings useful?
Paper • 2410.06205 • Published • 2 -
ThunderKittens: Simple, Fast, and Adorable AI Kernels
Paper • 2410.20399 • Published • 2
Core techniques: 1) unified grounding loss 2) language-aware deep fusion 3) pre-training with both types of data.
improved visual quality as the rough concept location and outline are decided in the early stages, followed by fine-grained details in later stages.
-
How Good Are Low-bit Quantized LLaMA3 Models? An Empirical Study
Paper • 2404.14047 • Published • 45 -
Reasoning in Large Language Models: A Geometric Perspective
Paper • 2407.02678 • Published • 1 -
Natural Language Reinforcement Learning
Paper • 2411.14251 • Published • 31 -
Byte Latent Transformer: Patches Scale Better Than Tokens
Paper • 2412.09871 • Published • 108
-
How Good Are Low-bit Quantized LLaMA3 Models? An Empirical Study
Paper • 2404.14047 • Published • 45 -
LiteSearch: Efficacious Tree Search for LLM
Paper • 2407.00320 • Published • 40 -
Cut Your Losses in Large-Vocabulary Language Models
Paper • 2411.09009 • Published • 49 -
LLaMA-Mesh: Unifying 3D Mesh Generation with Language Models
Paper • 2411.09595 • Published • 77
-
All you need is a good init
Paper • 1511.06422 • Published • 1 -
Align Your Steps: Optimizing Sampling Schedules in Diffusion Models
Paper • 2404.14507 • Published • 23 -
Efficient Transformer Encoders for Mask2Former-style models
Paper • 2404.15244 • Published • 1 -
Deep Residual Learning for Image Recognition
Paper • 1512.03385 • Published • 16
Encoder - Weighted Stochastic Depth
See Appendix A.2
-
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15 -
Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation
Paper • 2406.06525 • Published • 71 -
Data curation via joint example selection further accelerates multimodal learning
Paper • 2406.17711 • Published • 3 -
Unveiling Encoder-Free Vision-Language Models
Paper • 2406.11832 • Published • 55
-
Interactive3D: Create What You Want by Interactive 3D Generation
Paper • 2404.16510 • Published • 21 -
SEED-Bench-2-Plus: Benchmarking Multimodal Large Language Models with Text-Rich Visual Comprehension
Paper • 2404.16790 • Published • 10 -
A Thorough Examination of Decoding Methods in the Era of LLMs
Paper • 2402.06925 • Published • 1 -
LLaVA-OneVision: Easy Visual Task Transfer
Paper • 2408.03326 • Published • 61
-
SEED-Bench-2-Plus: Benchmarking Multimodal Large Language Models with Text-Rich Visual Comprehension
Paper • 2404.16790 • Published • 10 -
MMWorld: Towards Multi-discipline Multi-faceted World Model Evaluation in Videos
Paper • 2406.08407 • Published • 28 -
GMAI-MMBench: A Comprehensive Multimodal Evaluation Benchmark Towards General Medical AI
Paper • 2408.03361 • Published • 85
-
What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation
Paper • 2404.07129 • Published • 3 -
In-context Learning and Induction Heads
Paper • 2209.11895 • Published • 2 -
pyvene: A Library for Understanding and Improving PyTorch Models via Interventions
Paper • 2403.07809 • Published • 1
See: pattern-preserving ablation
Causal ablations taking into account LayerNorm
See figure 6: Classes vs labels in columns B and C. Subcircuit B delays phase change on number classes vs C delays on number of labels (dramatically)
See Appendix E: Merging Method Ablation on MoE Mistral and instruct 7B. Ties merged degenerate vs DARE model merges did not degenerate.
Local (gpu/cpu), compression (zstd) and over the wan (orc over s3 beats parquet too) results
See figure 19. Orc with cuDF beat parquet cuDF. Parquet arrow has dramatically more throughput without access to gpus
Additional paper with faq, code and tips on: https://github.com/microsoft/unilm/blob/master/bitnet/The-Era-of-1-bit-LLMs__Training_Tips_Code_FAQ.pdf
-
You Only Cache Once: Decoder-Decoder Architectures for Language Models
Paper • 2405.05254 • Published • 10 -
The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits
Paper • 2402.17764 • Published • 629 -
BitsFusion: 1.99 bits Weight Quantization of Diffusion Model
Paper • 2406.04333 • Published • 38 -
1-bit AI Infra: Part 1.1, Fast and Lossless BitNet b1.58 Inference on CPUs
Paper • 2410.16144 • Published • 5
Appendix A.9
Stage 2 training using LLaVA to describe the image with a caption
-
Adding Conditional Control to Text-to-Image Diffusion Models
Paper • 2302.05543 • Published • 58 -
Smoothed Energy Guidance: Guiding Diffusion Models with Reduced Energy Curvature of Attention
Paper • 2408.00760 • Published • 7 -
MagicQuill: An Intelligent Interactive Image Editing System
Paper • 2411.09703 • Published • 80 -
BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusion
Paper • 2403.06976 • Published • 2
-
QLoRA: Efficient Finetuning of Quantized LLMs
Paper • 2305.14314 • Published • 61 -
EfficientQAT: Efficient Quantization-Aware Training for Large Language Models
Paper • 2407.11062 • Published • 10 -
Spectra: A Comprehensive Study of Ternary, Quantized, and FP16 Language Models
Paper • 2407.12327 • Published • 79 -
BitNet a4.8: 4-bit Activations for 1-bit LLMs
Paper • 2411.04965 • Published • 69
Deterministic methods with unaligned models usually perform better on all tasks except for open-ended text generation.
-
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Efficient Algorithms for t-distributed Stochastic Neighborhood Embedding
Paper • 1712.09005 • Published • 1 -
UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
Paper • 1802.03426 • Published • 1 -
Principal subbundles for dimension reduction
Paper • 2307.03128 • Published • 1
-
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Barycentric Subspace Analysis on Manifolds
Paper • 1607.02833 • Published • 1 -
Template shape estimation: correcting an asymptotic bias
Paper • 1610.01502 • Published • 1 -
A Heat Diffusion Perspective on Geodesic Preserving Dimensionality Reduction
Paper • 2305.19043 • Published • 1
-
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Geodesic Multi-Modal Mixup for Robust Fine-Tuning
Paper • 2203.03897 • Published • 1 -
A Heat Diffusion Perspective on Geodesic Preserving Dimensionality Reduction
Paper • 2305.19043 • Published • 1 -
A micro Lie theory for state estimation in robotics
Paper • 1812.01537 • Published • 1
2008 paper: https://www.jmlr.org/papers/volume9/vandermaaten08a/vandermaaten08a.pdf
See also: https://dawn.cs.stanford.edu/2019/10/10/noneuclidean/
The manifold needs to be known for this layer to be implemented, and manifolds whose Log enjoys an analytical expression are preferred.
Euclidean signal on manifold domain for all inputs / outputs, with domain group action.
-
Architectures of Topological Deep Learning: A Survey on Topological Neural Networks
Paper • 2304.10031 • Published • 3 -
Adaptive Topological Feature via Persistent Homology: Filtration Learning for Point Clouds
Paper • 2307.09259 • Published • 1 -
Persistent homology of the cosmic web. I: Hierarchical topology in ΛCDM cosmologies
Paper • 2011.12851 • Published • 1
-
Fixup Initialization: Residual Learning Without Normalization
Paper • 1901.09321 • Published • 1 -
RegMixup: Mixup as a Regularizer Can Surprisingly Improve Accuracy and Out Distribution Robustness
Paper • 2206.14502 • Published • 1 -
MixUp as Locally Linear Out-Of-Manifold Regularization
Paper • 1809.02499 • Published • 1
https://github.com/clonebench/BigCloneBench and https://github.com/jeffsvajlenko/BigCloneEval
https://zenodo.org/records/4247595
https://pytorch.org/blog/flexattention/
Note: BPE does not handle incremental patching like BLT
Helps with finding the end of the byte patch
Patch start detected by entropy crossing a threshold
-
Qwen/QwQ-32B-Preview
Text Generation • 33B • Updated • 10.3k • • 1.74k -
bartowski/Qwen2.5-Coder-14B-Instruct-GGUF
Text Generation • 15B • Updated • 61.3k • 47 -
bartowski/Qwen2.5-Coder-32B-Instruct-GGUF
Text Generation • 33B • Updated • 36.5k • 106 -
bartowski/Qwen2.5-72B-Instruct-GGUF
Text Generation • 73B • Updated • 20.2k • 45
-
ByT5: Towards a token-free future with pre-trained byte-to-byte models
Paper • 2105.13626 • Published • 5 -
Byte Latent Transformer: Patches Scale Better Than Tokens
Paper • 2412.09871 • Published • 108 -
MrT5: Dynamic Token Merging for Efficient Byte-level Language Models
Paper • 2410.20771 • Published • 3
-
Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models
Paper • 2411.14257 • Published • 14 -
Scaling and evaluating sparse autoencoders
Paper • 2406.04093 • Published • 4 -
Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2
Paper • 2408.05147 • Published • 41 -
Disentangling Dense Embeddings with Sparse Autoencoders
Paper • 2408.00657 • Published • 1
-
Movie Gen: A Cast of Media Foundation Models
Paper • 2410.13720 • Published • 100 -
F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching
Paper • 2410.06885 • Published • 47 -
Flow Matching for Generative Modeling
Paper • 2210.02747 • Published • 4 -
Matcha-TTS: A fast TTS architecture with conditional flow matching
Paper • 2309.03199 • Published • 15
More extracted images on github: https://github.com/matlok-ai/python-copilot-image-and-audio-examples/tree/main/png
-
matlok/python-image-copilot-training-using-class-knowledge-graphs-2024-01-27
Viewer • Updated • 773 • 326 -
matlok/python-image-copilot-training-using-function-knowledge-graphs
Viewer • Updated • 88 • 1.41k -
matlok/python-image-copilot-training-using-inheritance-knowledge-graphs
Viewer • Updated • 88 • 1.38k -
matlok/python-image-copilot-training-using-import-knowledge-graphs
Viewer • Updated • 88 • 229
Python Alpaca instructions from leading AI research and tools repositories - focus is on "Manager level" understanding atm
-
matlok/python-text-copilot-training-instruct-ai-research-2024-02-11
Viewer • Updated • 130 • 1.79k -
matlok/python-text-copilot-training-instruct-ai-research-2024-02-10
Viewer • Updated • 123 • 4.21k -
matlok/python-text-copilot-training-instruct-ai-research-2024-02-03
Viewer • Updated • 2.67k • 646 • 1 -
matlok/python-text-copilot-training-instruct-ai-research-2024-01-27
Viewer • Updated • 43.1k • 159
More extracted mp3 samples on github: https://github.com/matlok-ai/python-copilot-image-and-audio-examples/tree/main/mp3
-
matlok/python-audio-copilot-training-using-class-knowledge-graphs-2024-01-27
Viewer • Updated • 948 • 567 -
matlok/python-audio-copilot-training-using-function-knowledge-graphs
Viewer • Updated • 120 • 1.92k • 1 -
matlok/python-audio-copilot-training-using-inheritance-knowledge-graphs
Viewer • Updated • 120 • 654 -
matlok/python-audio-copilot-training-using-import-knowledge-graphs
Viewer • Updated • 48 • 410
Python code from leading AI research and tools repositories
Great article on how this works: https://towardsdatascience.com/a-beginners-guide-to-llm-fine-tuning-4bae7d4da672
-
Visual Instruction Tuning
Paper • 2304.08485 • Published • 21 -
LLaVA-Plus: Learning to Use Tools for Creating Multimodal Agents
Paper • 2311.05437 • Published • 51 -
Improved Baselines with Visual Instruction Tuning
Paper • 2310.03744 • Published • 39 -
Aligning Large Multimodal Models with Factually Augmented RLHF
Paper • 2309.14525 • Published • 32
there's many more on arxiv if you search for CLAP
-
Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation
Paper • 2211.06687 • Published • 6 -
EnCLAP: Combining Neural Audio Codec and Audio-Text Joint Embedding for Automated Audio Captioning
Paper • 2401.17690 • Published • 5 -
Amphion: An Open-Source Audio, Music and Speech Generation Toolkit
Paper • 2312.09911 • Published • 55 -
Audiobox: Unified Audio Generation with Natural Language Prompts
Paper • 2312.15821 • Published • 16
-
Self-Instruct: Aligning Language Model with Self Generated Instructions
Paper • 2212.10560 • Published • 9 -
Principled Instructions Are All You Need for Questioning LLaMA-1/2, GPT-3.5/4
Paper • 2312.16171 • Published • 37 -
DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence
Paper • 2401.14196 • Published • 72 -
AlpaCare:Instruction-tuned Large Language Models for Medical Application
Paper • 2310.14558 • Published • 4
-
From Audio to Photoreal Embodiment: Synthesizing Humans in Conversations
Paper • 2401.01885 • Published • 28 -
Media2Face: Co-speech Facial Animation Generation With Multi-Modality Guidance
Paper • 2401.15687 • Published • 24 -
Unified-IO 2: Scaling Autoregressive Multimodal Models with Vision, Language, Audio, and Action
Paper • 2312.17172 • Published • 30 -
MouSi: Poly-Visual-Expert Vision-Language Models
Paper • 2401.17221 • Published • 9
MoE
-
MoE-LLaVA: Mixture of Experts for Large Vision-Language Models
Paper • 2401.15947 • Published • 53 -
DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models
Paper • 2401.06066 • Published • 61 -
SwitchHead: Accelerating Transformers with Mixture-of-Experts Attention
Paper • 2312.07987 • Published • 41 -
Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity
Paper • 2101.03961 • Published • 13
There's usually interesting papers in the model cards on the leaderboard: https://huggingface.co/spaces/bigcode/bigcode-models-leaderboard
-
StarCoder: may the source be with you!
Paper • 2305.06161 • Published • 33 -
WizardCoder: Empowering Code Large Language Models with Evol-Instruct
Paper • 2306.08568 • Published • 33 -
SantaCoder: don't reach for the stars!
Paper • 2301.03988 • Published • 7 -
DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence
Paper • 2401.14196 • Published • 72
-
dphn/dolphin-2.6-mistral-7b-dpo-laser
Text Generation • 7B • Updated • 77 • 120 -
Evaluating Large Language Models Trained on Code
Paper • 2107.03374 • Published • 10 -
CodeBERT: A Pre-Trained Model for Programming and Natural Languages
Paper • 2002.08155 • Published • 2 -
code2seq: Generating Sequences from Structured Representations of Code
Paper • 1808.01400 • Published • 2
-
Improving Text Embeddings with Large Language Models
Paper • 2401.00368 • Published • 83 -
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Paper • 1810.04805 • Published • 28 -
Metadata Might Make Language Models Better
Paper • 2211.10086 • Published • 4 -
DecoderLens: Layerwise Interpretation of Encoder-Decoder Transformers
Paper • 2310.03686 • Published • 3
Checkout: https://bbycroft.net/llm and http://nlp.seas.harvard.edu/2018/04/03/attention.html
-
Attention Is All You Need
Paper • 1706.03762 • Published • 122 -
ImageNet Large Scale Visual Recognition Challenge
Paper • 1409.0575 • Published • 10 -
Sequence to Sequence Learning with Neural Networks
Paper • 1409.3215 • Published • 3 -
Language Models are Few-Shot Learners
Paper • 2005.14165 • Published • 20
Large Multimodal Models
-
LCM-LoRA: A Universal Stable-Diffusion Acceleration Module
Paper • 2311.05556 • Published • 86 -
MultiLoRA: Democratizing LoRA for Better Multi-Task Learning
Paper • 2311.11501 • Published • 37 -
S-LoRA: Serving Thousands of Concurrent LoRA Adapters
Paper • 2311.03285 • Published • 30 -
LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70B
Paper • 2310.20624 • Published • 13
-
BLOOM: A 176B-Parameter Open-Access Multilingual Language Model
Paper • 2211.05100 • Published • 37 -
Contrastive Language-Image Pre-training for the Italian Language
Paper • 2108.08688 • Published • 2 -
IT5: Large-scale Text-to-text Pretraining for Italian Language Understanding and Generation
Paper • 2203.03759 • Published • 5 -
Spanish Pre-trained BERT Model and Evaluation Data
Paper • 2308.02976 • Published • 3
-
Metadata Might Make Language Models Better
Paper • 2211.10086 • Published • 4 -
Empirical Analysis of the Strengths and Weaknesses of PEFT Techniques for LLMs
Paper • 2304.14999 • Published • 2 -
PEFT for Speech: Unveiling Optimal Placement, Merging Strategies, and Ensemble Techniques
Paper • 2401.02122 • Published • 2 -
Zephyr: Direct Distillation of LM Alignment
Paper • 2310.16944 • Published • 123
-
Spam-T5: Benchmarking Large Language Models for Few-Shot Email Spam Detection
Paper • 2304.01238 • Published • 2 -
The FinBen: An Holistic Financial Benchmark for Large Language Models
Paper • 2402.12659 • Published • 24 -
TofuEval: Evaluating Hallucinations of LLMs on Topic-Focused Dialogue Summarization
Paper • 2402.13249 • Published • 15 -
Evaluating Large Language Models Trained on Code
Paper • 2107.03374 • Published • 10
-
metavoiceio/metavoice-1B-v0.1
Text-to-Speech • Updated • 39 • 790 -
BASE TTS: Lessons from building a billion-parameter Text-to-Speech model on 100K hours of data
Paper • 2402.08093 • Published • 61 -
EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions
Paper • 2402.17485 • Published • 194 -
SWivid/F5-TTS
Text-to-Speech • Updated • 482k • 1.17k
U-Net was trained in 10 hours on a NVidia Titan GPU (6 GB) - 2015
-
U-Net: Convolutional Networks for Biomedical Image Segmentation
Paper • 1505.04597 • Published • 18 -
microsoft/BioGPT-Large
Text Generation • Updated • 5.44k • 238 -
kuleshov-group/caduceus-ps_seqlen-131k_d_model-256_n_layer-16
Fill-Mask • 7.73M • Updated • 2.02k • 14 -
kuleshov-group/caduceus-ps_seqlen-1k_d_model-256_n_layer-4_lr-8e-3
Fill-Mask • 1.93M • Updated • 527 • 2
-
LAVE: LLM-Powered Agent Assistance and Language Augmentation for Video Editing
Paper • 2402.10294 • Published • 27 -
Valley: Video Assistant with Large Language model Enhanced abilitY
Paper • 2306.07207 • Published • 3 -
Video Editing via Factorized Diffusion Distillation
Paper • 2403.09334 • Published • 22
-
Linear Transformers with Learnable Kernel Functions are Better In-Context Models
Paper • 2402.10644 • Published • 81 -
GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
Paper • 2305.13245 • Published • 6 -
ChunkAttention: Efficient Self-Attention with Prefix-Aware KV Cache and Two-Phase Partition
Paper • 2402.15220 • Published • 20 -
Sequence Parallelism: Long Sequence Training from System Perspective
Paper • 2105.13120 • Published • 6
-
In Search of Needles in a 10M Haystack: Recurrent Memory Finds What LLMs Miss
Paper • 2402.10790 • Published • 42 -
LongAgent: Scaling Language Models to 128k Context through Multi-Agent Collaboration
Paper • 2402.11550 • Published • 19 -
A Neural Conversational Model
Paper • 1506.05869 • Published • 2 -
Data Engineering for Scaling Language Models to 128K Context
Paper • 2402.10171 • Published • 25
-
DataDreamer: A Tool for Synthetic Data Generation and Reproducible LLM Workflows
Paper • 2402.10379 • Published • 31 -
Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Grasping
Paper • 1709.07857 • Published • 2 -
Simple synthetic data reduces sycophancy in large language models
Paper • 2308.03958 • Published • 23 -
Qwen-VL: A Frontier Large Vision-Language Model with Versatile Abilities
Paper • 2308.12966 • Published • 12
-
FinTral: A Family of GPT-4 Level Multimodal Financial Large Language Models
Paper • 2402.10986 • Published • 83 -
BloombergGPT: A Large Language Model for Finance
Paper • 2303.17564 • Published • 32 -
GPT-InvestAR: Enhancing Stock Investment Strategies through Annual Report Analysis with Large Language Models
Paper • 2309.03079 • Published • 3 -
FinVis-GPT: A Multimodal Large Language Model for Financial Chart Analysis
Paper • 2308.01430 • Published • 2
-
AnyGPT: Unified Multimodal LLM with Discrete Sequence Modeling
Paper • 2402.12226 • Published • 45 -
M2-CLIP: A Multimodal, Multi-task Adapting Framework for Video Action Recognition
Paper • 2401.11649 • Published • 3 -
Gen4Gen: Generative Data Pipeline for Generative Multi-Concept Composition
Paper • 2402.15504 • Published • 21 -
EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions
Paper • 2402.17485 • Published • 194
-
A Generalized Language Model as the Combination of Skipped n-grams and Modified Kneser-Ney Smoothing
Paper • 1404.3377 • Published • 2 -
Skip-gram Language Modeling Using Sparse Non-negative Matrix Probability Estimation
Paper • 1412.1454 • Published • 2 -
Neural Text Generation from Structured Data with Application to the Biography Domain
Paper • 1603.07771 • Published • 2 -
Distributed Representations of Words and Phrases and their Compositionality
Paper • 1310.4546 • Published • 3
-
VideoPrism: A Foundational Visual Encoder for Video Understanding
Paper • 2402.13217 • Published • 40 -
EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions
Paper • 2402.17485 • Published • 194 -
Qwen/Qwen-VL-Chat
Text Generation • Updated • 89.3k • 383 -
MovieLLM: Enhancing Long Video Understanding with AI-Generated Movies
Paper • 2403.01422 • Published • 30
-
Improving Robustness for Joint Optimization of Camera Poses and Decomposed Low-Rank Tensorial Radiance Fields
Paper • 2402.13252 • Published • 19 -
RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputs
Paper • 2112.00724 • Published • 2 -
Enhancing NeRF akin to Enhancing LLMs: Generalizable NeRF Transformer with Mixture-of-View-Experts
Paper • 2308.11793 • Published • 2
-
Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Grasping
Paper • 1709.07857 • Published • 2 -
Sensor-based Multi-Robot Search and Coverage with Spatial Separation in Unstructured Environments
Paper • 2403.01710 • Published • 2 -
Twisting Lids Off with Two Hands
Paper • 2403.02338 • Published • 7
-
Training Compute-Optimal Large Language Models
Paper • 2203.15556 • Published • 11 -
Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
Paper • 1909.08053 • Published • 5 -
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Paper • 1910.10683 • Published • 18 -
Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling
Paper • 2304.01373 • Published • 9
exploring speculative sampling with autoregressive model like: https://proceedings.mlr.press/v139/song21a.html and https://proceedings.mlr.press/v119/
-
Lossless Acceleration for Seq2seq Generation with Aggressive Decoding
Paper • 2205.10350 • Published • 2 -
Blockwise Parallel Decoding for Deep Autoregressive Models
Paper • 1811.03115 • Published • 2 -
Fast Transformer Decoding: One Write-Head is All You Need
Paper • 1911.02150 • Published • 9 -
Sequence-Level Knowledge Distillation
Paper • 1606.07947 • Published • 2
-
CodeBERT: A Pre-Trained Model for Programming and Natural Languages
Paper • 2002.08155 • Published • 2 -
OpenCodeInterpreter: Integrating Code Generation with Execution and Refinement
Paper • 2402.14658 • Published • 84 -
CodeFusion: A Pre-trained Diffusion Model for Code Generation
Paper • 2310.17680 • Published • 74 -
CodePlan: Repository-level Coding using LLMs and Planning
Paper • 2309.12499 • Published • 80
-
Orca-Math: Unlocking the potential of SLMs in Grade School Math
Paper • 2402.14830 • Published • 24 -
MathScale: Scaling Instruction Tuning for Mathematical Reasoning
Paper • 2403.02884 • Published • 17 -
meta-math/MetaMath-Mistral-7B
Text Generation • Updated • 1.26k • • 96 -
meta-math/MetaMath-13B-V1.0
Text Generation • Updated • 1k • 13
-
Same Task, More Tokens: the Impact of Input Length on the Reasoning Performance of Large Language Models
Paper • 2402.14848 • Published • 19 -
Teaching Large Language Models to Reason with Reinforcement Learning
Paper • 2403.04642 • Published • 48 -
How Far Are We from Intelligent Visual Deductive Reasoning?
Paper • 2403.04732 • Published • 21 -
Learning to Reason and Memorize with Self-Notes
Paper • 2305.00833 • Published • 5
-
MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases
Paper • 2402.14905 • Published • 134 -
Sensor-based Multi-Robot Search and Coverage with Spatial Separation in Unstructured Environments
Paper • 2403.01710 • Published • 2 -
EdgeMoE: Fast On-Device Inference of MoE-based Large Language Models
Paper • 2308.14352 • Published -
Slimmable Encoders for Flexible Split DNNs in Bandwidth and Resource Constrained IoT Systems
Paper • 2306.12691 • Published • 3
-
Geometric Algebra Transformers
Paper • 2305.18415 • Published • 3 -
World Model on Million-Length Video And Language With RingAttention
Paper • 2402.08268 • Published • 40 -
Deep Unsupervised Learning using Nonequilibrium Thermodynamics
Paper • 1503.03585 • Published • 6 -
IDKiro/sdxs-512-0.9
Text-to-Image • Updated • 807 • 110
-
Video as the New Language for Real-World Decision Making
Paper • 2402.17139 • Published • 22 -
VideoCrafter1: Open Diffusion Models for High-Quality Video Generation
Paper • 2310.19512 • Published • 16 -
VideoMamba: State Space Model for Efficient Video Understanding
Paper • 2403.06977 • Published • 29 -
VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models
Paper • 2401.09047 • Published • 14
-
introspector/unimath
Updated • 3.04k • 7 -
MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms
Paper • 1905.13319 • Published • 2 -
Measuring Mathematical Problem Solving With the MATH Dataset
Paper • 2103.03874 • Published • 6 -
MathScale: Scaling Instruction Tuning for Mathematical Reasoning
Paper • 2403.02884 • Published • 17
-
Priority Sampling of Large Language Models for Compilers
Paper • 2402.18734 • Published • 19 -
Accelerating Large Language Model Decoding with Speculative Sampling
Paper • 2302.01318 • Published • 4 -
Fast Inference from Transformers via Speculative Decoding
Paper • 2211.17192 • Published • 11 -
AttentiveNAS: Improving Neural Architecture Search via Attentive Sampling
Paper • 2011.09011 • Published • 2
-
Resonance RoPE: Improving Context Length Generalization of Large Language Models
Paper • 2403.00071 • Published • 24 -
Scaling Laws of RoPE-based Extrapolation
Paper • 2310.05209 • Published • 8 -
Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models
Paper • 2404.12387 • Published • 40 -
OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework
Paper • 2404.14619 • Published • 126
-
Training Verifiers to Solve Math Word Problems
Paper • 2110.14168 • Published • 7 -
MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
Paper • 2309.12284 • Published • 19 -
LiteSearch: Efficacious Tree Search for LLM
Paper • 2407.00320 • Published • 40 -
DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models
Paper • 2309.03883 • Published • 36
-
Scaling Laws for Neural Language Models
Paper • 2001.08361 • Published • 10 -
An Empirical Model of Large-Batch Training
Paper • 1812.06162 • Published • 3 -
Measuring the Effects of Data Parallelism on Neural Network Training
Paper • 1811.03600 • Published • 2 -
Adafactor: Adaptive Learning Rates with Sublinear Memory Cost
Paper • 1804.04235 • Published • 2
-
Measuring the Effects of Data Parallelism on Neural Network Training
Paper • 1811.03600 • Published • 2 -
Adafactor: Adaptive Learning Rates with Sublinear Memory Cost
Paper • 1804.04235 • Published • 2 -
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
Paper • 1905.11946 • Published • 4 -
Yi: Open Foundation Models by 01.AI
Paper • 2403.04652 • Published • 65
-
Sequence Parallelism: Long Sequence Training from System Perspective
Paper • 2105.13120 • Published • 6 -
Ring Attention with Blockwise Transformers for Near-Infinite Context
Paper • 2310.01889 • Published • 13 -
Striped Attention: Faster Ring Attention for Causal Transformers
Paper • 2311.09431 • Published • 4 -
World Model on Million-Length Video And Language With RingAttention
Paper • 2402.08268 • Published • 40
-
Accelerating LLM Inference with Staged Speculative Decoding
Paper • 2308.04623 • Published • 26 -
An Emulator for Fine-Tuning Large Language Models using Small Language Models
Paper • 2310.12962 • Published • 13 -
The Curious Case of Neural Text Degeneration
Paper • 1904.09751 • Published • 3 -
On Speculative Decoding for Multimodal Large Language Models
Paper • 2404.08856 • Published • 13
-
ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment
Paper • 2403.05135 • Published • 45 -
DeepSeek-VL: Towards Real-World Vision-Language Understanding
Paper • 2403.05525 • Published • 49 -
CoCa: Contrastive Captioners are Image-Text Foundation Models
Paper • 2205.01917 • Published • 3
-
Mixtral of Experts
Paper • 2401.04088 • Published • 160 -
MoE-LLaVA: Mixture of Experts for Large Vision-Language Models
Paper • 2401.15947 • Published • 53 -
MoE-Mamba: Efficient Selective State Space Models with Mixture of Experts
Paper • 2401.04081 • Published • 74 -
EdgeMoE: Fast On-Device Inference of MoE-based Large Language Models
Paper • 2308.14352 • Published
-
Non-asymptotic oracle inequalities for the Lasso in high-dimensional mixture of experts
Paper • 2009.10622 • Published • 1 -
MoE-LLaVA: Mixture of Experts for Large Vision-Language Models
Paper • 2401.15947 • Published • 53 -
MoE-Mamba: Efficient Selective State Space Models with Mixture of Experts
Paper • 2401.04081 • Published • 74 -
MoE-Infinity: Activation-Aware Expert Offloading for Efficient MoE Serving
Paper • 2401.14361 • Published • 2
-
EVE: Efficient Vision-Language Pre-training with Masked Prediction and Modality-Aware MoE
Paper • 2308.11971 • Published • 2 -
HyperFormer: Enhancing Entity and Relation Interaction for Hyper-Relational Knowledge Graph Completion
Paper • 2308.06512 • Published • 2 -
Unraveling Complex Data Diversity in Underwater Acoustic Target Recognition through Convolution-based Mixture of Experts
Paper • 2402.11919 • Published • 2
-
Adaptive sequential Monte Carlo by means of mixture of experts
Paper • 1108.2836 • Published • 2 -
Convergence Rates for Mixture-of-Experts
Paper • 1110.2058 • Published • 2 -
Multi-view Contrastive Learning for Entity Typing over Knowledge Graphs
Paper • 2310.12008 • Published • 2 -
Enhancing NeRF akin to Enhancing LLMs: Generalizable NeRF Transformer with Mixture-of-View-Experts
Paper • 2308.11793 • Published • 2
-
Multi-view Contrastive Learning for Entity Typing over Knowledge Graphs
Paper • 2310.12008 • Published • 2 -
HyperFormer: Enhancing Entity and Relation Interaction for Hyper-Relational Knowledge Graph Completion
Paper • 2308.06512 • Published • 2 -
ARIEL: Adversarial Graph Contrastive Learning
Paper • 2208.06956 • Published • 2 -
RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval
Paper • 2401.18059 • Published • 48
-
Robust Mixture-of-Expert Training for Convolutional Neural Networks
Paper • 2308.10110 • Published • 2 -
Experts Weights Averaging: A New General Training Scheme for Vision Transformers
Paper • 2308.06093 • Published • 2 -
ConstitutionalExperts: Training a Mixture of Principle-based Prompts
Paper • 2403.04894 • Published • 2 -
Mixture-of-LoRAs: An Efficient Multitask Tuning for Large Language Models
Paper • 2403.03432 • Published • 1
-
Robust Mixture-of-Expert Training for Convolutional Neural Networks
Paper • 2308.10110 • Published • 2 -
HyperFormer: Enhancing Entity and Relation Interaction for Hyper-Relational Knowledge Graph Completion
Paper • 2308.06512 • Published • 2 -
Mobile V-MoEs: Scaling Down Vision Transformers via Sparse Mixture-of-Experts
Paper • 2309.04354 • Published • 16 -
Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints
Paper • 2212.05055 • Published • 6
-
Mixture-of-LoRAs: An Efficient Multitask Tuning for Large Language Models
Paper • 2403.03432 • Published • 1 -
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data
Paper • 2304.08247 • Published • 2
-
Turn Waste into Worth: Rectifying Top-k Router of MoE
Paper • 2402.12399 • Published • 2 -
CompeteSMoE -- Effective Training of Sparse Mixture of Experts via Competition
Paper • 2402.02526 • Published • 3 -
Buffer Overflow in Mixture of Experts
Paper • 2402.05526 • Published • 9 -
OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models
Paper • 2402.01739 • Published • 28
-
Scaling Vision with Sparse Mixture of Experts
Paper • 2106.05974 • Published • 4 -
Routers in Vision Mixture of Experts: An Empirical Study
Paper • 2401.15969 • Published • 2 -
Multimodal Contrastive Learning with LIMoE: the Language-Image Mixture of Experts
Paper • 2206.02770 • Published • 4 -
Experts Weights Averaging: A New General Training Scheme for Vision Transformers
Paper • 2308.06093 • Published • 2
-
Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
Paper • 1701.06538 • Published • 7 -
ST-MoE: Designing Stable and Transferable Sparse Expert Models
Paper • 2202.08906 • Published • 3 -
Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM
Paper • 2403.07816 • Published • 45
-
LocMoE: A Low-overhead MoE for Large Language Model Training
Paper • 2401.13920 • Published • 2 -
HyperRouter: Towards Efficient Training and Inference of Sparse Mixture of Experts
Paper • 2312.07035 • Published • 2 -
DEMix Layers: Disentangling Domains for Modular Language Modeling
Paper • 2108.05036 • Published • 3
-
TinyLLaVA: A Framework of Small-scale Large Multimodal Models
Paper • 2402.14289 • Published • 20 -
ImageBind: One Embedding Space To Bind Them All
Paper • 2305.05665 • Published • 6 -
DocLLM: A layout-aware generative language model for multimodal document understanding
Paper • 2401.00908 • Published • 191 -
Multimodal Contrastive Learning with LIMoE: the Language-Image Mixture of Experts
Paper • 2206.02770 • Published • 4
-
DocLLM: A layout-aware generative language model for multimodal document understanding
Paper • 2401.00908 • Published • 191 -
Unifying Vision, Text, and Layout for Universal Document Processing
Paper • 2212.02623 • Published • 12 -
Grounded Language-Image Pre-training
Paper • 2112.03857 • Published • 3 -
ConsistencyDet: Robust Object Detector with Denoising Paradigm of Consistency Model
Paper • 2404.07773 • Published • 1
-
DocLLM: A layout-aware generative language model for multimodal document understanding
Paper • 2401.00908 • Published • 191 -
DeBERTa: Decoding-enhanced BERT with Disentangled Attention
Paper • 2006.03654 • Published • 3 -
DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing
Paper • 2111.09543 • Published • 3
-
FaceChain-SuDe: Building Derived Class to Inherit Category Attributes for One-shot Subject-Driven Generation
Paper • 2403.06775 • Published • 4 -
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Paper • 2010.11929 • Published • 15 -
Data Incubation -- Synthesizing Missing Data for Handwriting Recognition
Paper • 2110.07040 • Published • 2 -
A Mixture of Expert Approach for Low-Cost Customization of Deep Neural Networks
Paper • 1811.00056 • Published • 2
-
Adding NVMe SSDs to Enable and Accelerate 100B Model Fine-tuning on a Single GPU
Paper • 2403.06504 • Published • 56 -
Token-Level Adaptation of LoRA Adapters for Downstream Task Generalization
Paper • 2311.10847 • Published • 2 -
PERL: Parameter Efficient Reinforcement Learning from Human Feedback
Paper • 2403.10704 • Published • 60
-
SpeechMoE: Scaling to Large Acoustic Models with Dynamic Routing Mixture of Experts
Paper • 2105.03036 • Published • 2 -
Building a great multi-lingual teacher with sparsely-gated mixture of experts for speech recognition
Paper • 2112.05820 • Published • 2 -
SpeechMoE2: Mixture-of-Experts Model with Improved Routing
Paper • 2111.11831 • Published • 2
-
QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models
Paper • 2310.16795 • Published • 27 -
Pareto-Optimal Quantized ResNet Is Mostly 4-bit
Paper • 2105.03536 • Published • 3 -
Decoding Compressed Trust: Scrutinizing the Trustworthiness of Efficient LLMs Under Compression
Paper • 2403.15447 • Published • 16
-
Mixture-of-Supernets: Improving Weight-Sharing Supernet Training with Architecture-Routed Mixture-of-Experts
Paper • 2306.04845 • Published • 4 -
Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networks
Paper • 2306.04073 • Published • 2 -
Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM
Paper • 2403.07816 • Published • 45 -
Unified Scaling Laws for Routed Language Models
Paper • 2202.01169 • Published • 2
-
Data Incubation -- Synthesizing Missing Data for Handwriting Recognition
Paper • 2110.07040 • Published • 2 -
A Mixture of Expert Approach for Low-Cost Customization of Deep Neural Networks
Paper • 1811.00056 • Published • 2 -
Vulnerability Analysis of Transformer-based Optical Character Recognition to Adversarial Attacks
Paper • 2311.17128 • Published • 2 -
Data Generation for Post-OCR correction of Cyrillic handwriting
Paper • 2311.15896 • Published • 4
-
Vulnerability Analysis of Transformer-based Optical Character Recognition to Adversarial Attacks
Paper • 2311.17128 • Published • 2 -
Data Generation for Post-OCR correction of Cyrillic handwriting
Paper • 2311.15896 • Published • 4 -
An End-to-End OCR Framework for Robust Arabic-Handwriting Recognition using a Novel Transformers-based Model and an Innovative 270 Million-Words Multi-Font Corpus of Classical Arabic with Diacritics
Paper • 2208.11484 • Published • 3 -
Transformer based Urdu Handwritten Text Optical Character Reader
Paper • 2206.04575 • Published • 2
-
Character Queries: A Transformer-based Approach to On-Line Handwritten Character Segmentation
Paper • 2309.03072 • Published • 2 -
Prompt me a Dataset: An investigation of text-image prompting for historical image dataset creation using foundation models
Paper • 2309.01674 • Published • 2 -
Segment Anything
Paper • 2304.02643 • Published • 6
-
Disentangling Writer and Character Styles for Handwriting Generation
Paper • 2303.14736 • Published • 3 -
A Transformer Architecture for Online Gesture Recognition of Mathematical Expressions
Paper • 2211.02643 • Published • 2 -
A tailored Handwritten-Text-Recognition System for Medieval Latin
Paper • 2308.09368 • Published • 3 -
Scalable handwritten text recognition system for lexicographic sources of under-resourced languages and alphabets
Paper • 2303.16256 • Published • 2
-
DocLLM: A layout-aware generative language model for multimodal document understanding
Paper • 2401.00908 • Published • 191 -
Visual Instruction Tuning
Paper • 2304.08485 • Published • 21 -
Glyph-ByT5: A Customized Text Encoder for Accurate Visual Text Rendering
Paper • 2403.09622 • Published • 17 -
Lumiere: A Space-Time Diffusion Model for Video Generation
Paper • 2401.12945 • Published • 86
GitHub: https://github.com/Planet-AI-GmbH/tfaip-hybrid-ctc-s2s and math: https://storage.googleapis.com/mathwriting_data/mathwriting-2024.tgz
-
Rescoring Sequence-to-Sequence Models for Text Line Recognition with CTC-Prefixes
Paper • 2110.05909 • Published • 2 -
Deep Residual Learning for Image Recognition
Paper • 1512.03385 • Published • 16 -
Wide Residual Networks
Paper • 1605.07146 • Published • 2 -
Comprehensive Survey of Model Compression and Speed up for Vision Transformers
Paper • 2404.10407 • Published • 1
-
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Paper • 1810.04805 • Published • 28 -
Transformers Can Achieve Length Generalization But Not Robustly
Paper • 2402.09371 • Published • 14 -
Triple-Encoders: Representations That Fire Together, Wire Together
Paper • 2402.12332 • Published • 2 -
BERTs are Generative In-Context Learners
Paper • 2406.04823 • Published • 1
-
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Paper • 1810.04805 • Published • 28 -
Transformers Can Achieve Length Generalization But Not Robustly
Paper • 2402.09371 • Published • 14 -
A Thorough Examination of Decoding Methods in the Era of LLMs
Paper • 2402.06925 • Published • 1 -
Byte Latent Transformer: Patches Scale Better Than Tokens
Paper • 2412.09871 • Published • 108
-
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
Paper • 1901.08746 • Published • 6 -
Pretraining-Based Natural Language Generation for Text Summarization
Paper • 1902.09243 • Published • 2 -
RoBERTa: A Robustly Optimized BERT Pretraining Approach
Paper • 1907.11692 • Published • 10 -
DeBERTa: Decoding-enhanced BERT with Disentangled Attention
Paper • 2006.03654 • Published • 3
-
Pretraining-Based Natural Language Generation for Text Summarization
Paper • 1902.09243 • Published • 2 -
Unified Vision-Language Pre-Training for Image Captioning and VQA
Paper • 1909.11059 • Published • 2 -
When LLMs are Unfit Use FastFit: Fast and Effective Text Classification with Many Classes
Paper • 2404.12365 • Published • 2
-
Pretraining-Based Natural Language Generation for Text Summarization
Paper • 1902.09243 • Published • 2 -
Learning to Reason and Memorize with Self-Notes
Paper • 2305.00833 • Published • 5 -
Text Generation with Diffusion Language Models: A Pre-training Approach with Continuous Paragraph Denoise
Paper • 2212.11685 • Published • 2 -
Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems
Paper • 2408.16293 • Published • 27
-
Representing Online Handwriting for Recognition in Large Vision-Language Models
Paper • 2402.15307 • Published • 3 -
Evaluating Sequence-to-Sequence Models for Handwritten Text Recognition
Paper • 1903.07377 • Published • 2 -
Enhancing Document Information Analysis with Multi-Task Pre-training: A Robust Approach for Information Extraction in Visually-Rich Documents
Paper • 2310.16527 • Published • 2 -
Detecting and recognizing characters in Greek papyri with YOLOv8, DeiT and SimCLR
Paper • 2401.12513 • Published • 1
-
An Interdisciplinary Comparison of Sequence Modeling Methods for Next-Element Prediction
Paper • 1811.00062 • Published • 2 -
mT5: A massively multilingual pre-trained text-to-text transformer
Paper • 2010.11934 • Published • 4 -
Bootstrap Your Own Skills: Learning to Solve New Tasks with Large Language Model Guidance
Paper • 2310.10021 • Published • 2 -
Gemma: Open Models Based on Gemini Research and Technology
Paper • 2403.08295 • Published • 50
-
mT5: A massively multilingual pre-trained text-to-text transformer
Paper • 2010.11934 • Published • 4 -
mSLAM: Massively multilingual joint pre-training for speech and text
Paper • 2202.01374 • Published • 2 -
DeepNet: Scaling Transformers to 1,000 Layers
Paper • 2203.00555 • Published • 2 -
SONAR: Sentence-Level Multimodal and Language-Agnostic Representations
Paper • 2308.11466 • Published • 1
-
Enhancing Document Information Analysis with Multi-Task Pre-training: A Robust Approach for Information Extraction in Visually-Rich Documents
Paper • 2310.16527 • Published • 2 -
CoDA: Collaborative Novel Box Discovery and Cross-modal Alignment for Open-vocabulary 3D Object Detection
Paper • 2310.02960 • Published • 2 -
MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training
Paper • 2403.09611 • Published • 129 -
Veagle: Advancements in Multimodal Representation Learning
Paper • 2403.08773 • Published • 10
-
Enhancing Document Information Analysis with Multi-Task Pre-training: A Robust Approach for Information Extraction in Visually-Rich Documents
Paper • 2310.16527 • Published • 2 -
DocLLM: A layout-aware generative language model for multimodal document understanding
Paper • 2401.00908 • Published • 191 -
Unifying Vision, Text, and Layout for Universal Document Processing
Paper • 2212.02623 • Published • 12
-
DragAnything: Motion Control for Anything using Entity Representation
Paper • 2403.07420 • Published • 14 -
Capabilities of Gemini Models in Medicine
Paper • 2404.18416 • Published • 25 -
MMWorld: Towards Multi-discipline Multi-faceted World Model Evaluation in Videos
Paper • 2406.08407 • Published • 28
-
Synth^2: Boosting Visual-Language Models with Synthetic Captions and Image Embeddings
Paper • 2403.07750 • Published • 23 -
Gaussian Frosting: Editable Complex Radiance Fields with Real-Time Rendering
Paper • 2403.14554 • Published • 14 -
ObjectDrop: Bootstrapping Counterfactuals for Photorealistic Object Removal and Insertion
Paper • 2403.18818 • Published • 28
-
Wide Residual Networks
Paper • 1605.07146 • Published • 2 -
Characterizing signal propagation to close the performance gap in unnormalized ResNets
Paper • 2101.08692 • Published • 2 -
Pareto-Optimal Quantized ResNet Is Mostly 4-bit
Paper • 2105.03536 • Published • 3 -
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations
Paper • 2106.01548 • Published • 2
-
Mixture-of-Supernets: Improving Weight-Sharing Supernet Training with Architecture-Routed Mixture-of-Experts
Paper • 2306.04845 • Published • 4 -
Balanced Mixture of SuperNets for Learning the CNN Pooling Architecture
Paper • 2306.11982 • Published • 2 -
AlphaNet: Improved Training of Supernets with Alpha-Divergence
Paper • 2102.07954 • Published • 2
-
Motion Mamba: Efficient and Long Sequence Motion Generation with Hierarchical and Bidirectional Selective SSM
Paper • 2403.07487 • Published • 16 -
LocalMamba: Visual State Space Model with Windowed Selective Scan
Paper • 2403.09338 • Published • 8 -
Cobra: Extending Mamba to Multi-Modal Large Language Model for Efficient Inference
Paper • 2403.14520 • Published • 35 -
SiMBA: Simplified Mamba-Based Architecture for Vision and Multivariate Time series
Paper • 2403.15360 • Published • 13
Mixtures of experts for text, image and speech
-
Chronos: Learning the Language of Time Series
Paper • 2403.07815 • Published • 50 -
Large Language Models as Optimizers
Paper • 2309.03409 • Published • 79 -
Pattern Discovery in Time Series with Byte Pair Encoding
Paper • 2106.00614 • Published • 2 -
MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection
Paper • 2403.19888 • Published • 12
-
AttentiveNAS: Improving Neural Architecture Search via Attentive Sampling
Paper • 2011.09011 • Published • 2 -
HAT: Hardware-Aware Transformers for Efficient Natural Language Processing
Paper • 2005.14187 • Published • 2 -
BigNAS: Scaling Up Neural Architecture Search with Big Single-Stage Models
Paper • 2003.11142 • Published • 2 -
Efficient Architecture Search by Network Transformation
Paper • 1707.04873 • Published • 2
-
U-Net: Convolutional Networks for Biomedical Image Segmentation
Paper • 1505.04597 • Published • 18 -
Image Segmentation using U-Net Architecture for Powder X-ray Diffraction Images
Paper • 2310.16186 • Published • 2 -
H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes
Paper • 1709.07330 • Published • 2 -
Deep LOGISMOS: Deep Learning Graph-based 3D Segmentation of Pancreatic Tumors on CT scans
Paper • 1801.08599 • Published • 2
-
Image Segmentation using U-Net Architecture for Powder X-ray Diffraction Images
Paper • 2310.16186 • Published • 2 -
H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes
Paper • 1709.07330 • Published • 2 -
Deep LOGISMOS: Deep Learning Graph-based 3D Segmentation of Pancreatic Tumors on CT scans
Paper • 1801.08599 • Published • 2 -
RTSeg: Real-time Semantic Segmentation Comparative Study
Paper • 1803.02758 • Published • 2
-
H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes
Paper • 1709.07330 • Published • 2 -
Deep LOGISMOS: Deep Learning Graph-based 3D Segmentation of Pancreatic Tumors on CT scans
Paper • 1801.08599 • Published • 2 -
Hierarchical multi-class segmentation of glioma images using networks with multi-level activation function
Paper • 1810.09488 • Published • 2 -
Cross-modality (CT-MRI) prior augmented deep learning for robust lung tumor segmentation from small MR datasets
Paper • 1901.11369 • Published • 2
-
MovieLLM: Enhancing Long Video Understanding with AI-Generated Movies
Paper • 2403.01422 • Published • 30 -
VisionGPT-3D: A Generalized Multimodal Agent for Enhanced 3D Vision Understanding
Paper • 2403.09530 • Published • 10 -
VidToMe: Video Token Merging for Zero-Shot Video Editing
Paper • 2312.10656 • Published • 11 -
TC4D: Trajectory-Conditioned Text-to-4D Generation
Paper • 2403.17920 • Published • 18
-
Generalizability vs. Robustness: Adversarial Examples for Medical Imaging
Paper • 1804.00504 • Published • 2 -
Evaluating Transformer-based Semantic Segmentation Networks for Pathological Image Segmentation
Paper • 2108.11993 • Published • 2 -
From Modern CNNs to Vision Transformers: Assessing the Performance, Robustness, and Classification Strategies of Deep Learning Models in Histopathology
Paper • 2204.05044 • Published • 2
-
3D Medical Image Segmentation based on multi-scale MPU-Net
Paper • 2307.05799 • Published • 2 -
Joint Liver and Hepatic Lesion Segmentation in MRI using a Hybrid CNN with Transformer Layers
Paper • 2201.10981 • Published • 2 -
Using Multi-scale SwinTransformer-HTC with Data augmentation in CoNIC Challenge
Paper • 2202.13588 • Published • 2
-
Attention Swin U-Net: Cross-Contextual Attention Mechanism for Skin Lesion Segmentation
Paper • 2210.16898 • Published • 2 -
Cross-Shaped Windows Transformer with Self-supervised Pretraining for Clinically Significant Prostate Cancer Detection in Bi-parametric MRI
Paper • 2305.00385 • Published • 2 -
Event Camera Demosaicing via Swin Transformer and Pixel-focus Loss
Paper • 2404.02731 • Published • 1
-
Enforcing Morphological Information in Fully Convolutional Networks to Improve Cell Instance Segmentation in Fluorescence Microscopy Images
Paper • 2106.05843 • Published • 2 -
Semi-Supervised Semantic Segmentation using Redesigned Self-Training for White Blood Cells
Paper • 2401.07278 • Published • 2
-
CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows
Paper • 2107.00652 • Published • 2 -
Cross-Shaped Windows Transformer with Self-supervised Pretraining for Clinically Significant Prostate Cancer Detection in Bi-parametric MRI
Paper • 2305.00385 • Published • 2 -
2nd Place Solution to Google Landmark Recognition Competition 2021
Paper • 2110.02638 • Published • 2 -
BOAT: Bilateral Local Attention Vision Transformer
Paper • 2201.13027 • Published • 2
-
CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows
Paper • 2107.00652 • Published • 2 -
Glyph-ByT5: A Customized Text Encoder for Accurate Visual Text Rendering
Paper • 2403.09622 • Published • 17 -
Veagle: Advancements in Multimodal Representation Learning
Paper • 2403.08773 • Published • 10 -
mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality
Paper • 2304.14178 • Published • 3
-
MAFormer: A Transformer Network with Multi-scale Attention Fusion for Visual Recognition
Paper • 2209.01620 • Published • 2 -
Using Multi-scale SwinTransformer-HTC with Data augmentation in CoNIC Challenge
Paper • 2202.13588 • Published • 2 -
GasHis-Transformer: A Multi-scale Visual Transformer Approach for Gastric Histopathological Image Detection
Paper • 2104.14528 • Published • 2
-
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
Paper • 2103.14030 • Published • 5 -
A Novel Transformer Based Semantic Segmentation Scheme for Fine-Resolution Remote Sensing Images
Paper • 2104.12137 • Published • 2 -
Self-Supervised Learning with Swin Transformers
Paper • 2105.04553 • Published • 3 -
Evaluating Transformer-based Semantic Segmentation Networks for Pathological Image Segmentation
Paper • 2108.11993 • Published • 2
-
Bootstrap Your Own Skills: Learning to Solve New Tasks with Large Language Model Guidance
Paper • 2310.10021 • Published • 2 -
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
Discovering Adaptable Symbolic Algorithms from Scratch
Paper • 2307.16890 • Published • 8 -
DragAPart: Learning a Part-Level Motion Prior for Articulated Objects
Paper • 2403.15382 • Published • 11
-
Self-Supervised Vision Transformers Learn Visual Concepts in Histopathology
Paper • 2203.00585 • Published • 3 -
Emerging Properties in Self-Supervised Vision Transformers
Paper • 2104.14294 • Published • 4 -
DreamScene360: Unconstrained Text-to-3D Scene Generation with Panoramic Gaussian Splatting
Paper • 2404.06903 • Published • 21 -
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32
-
Feature-Guided Black-Box Safety Testing of Deep Neural Networks
Paper • 1710.07859 • Published • 2 -
Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks
Paper • 2309.17410 • Published • 4 -
Intriguing Properties of Adversarial Examples
Paper • 1711.02846 • Published • 2
-
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
JoMA: Demystifying Multilayer Transformers via JOint Dynamics of MLP and Attention
Paper • 2310.00535 • Published • 2 -
Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla
Paper • 2307.09458 • Published • 12 -
The Impact of Depth and Width on Transformer Language Model Generalization
Paper • 2310.19956 • Published • 10
-
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
Chain-of-Thought Reasoning Without Prompting
Paper • 2402.10200 • Published • 110 -
LLM2LLM: Boosting LLMs with Novel Iterative Data Enhancement
Paper • 2403.15042 • Published • 27 -
LIMA: Less Is More for Alignment
Paper • 2305.11206 • Published • 27
-
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
SELF: Language-Driven Self-Evolution for Large Language Model
Paper • 2310.00533 • Published • 2 -
QLoRA: Efficient Finetuning of Quantized LLMs
Paper • 2305.14314 • Published • 61 -
QA-LoRA: Quantization-Aware Low-Rank Adaptation of Large Language Models
Paper • 2309.14717 • Published • 46
-
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data
Paper • 2304.08247 • Published • 2 -
S-LoRA: Serving Thousands of Concurrent LoRA Adapters
Paper • 2311.03285 • Published • 30 -
WavLLM: Towards Robust and Adaptive Speech Large Language Model
Paper • 2404.00656 • Published • 11
-
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
SELF: Language-Driven Self-Evolution for Large Language Model
Paper • 2310.00533 • Published • 2 -
Bigger, Better, Faster: Human-level Atari with human-level efficiency
Paper • 2305.19452 • Published • 5 -
DeepSeek-Prover-V1.5: Harnessing Proof Assistant Feedback for Reinforcement Learning and Monte-Carlo Tree Search
Paper • 2408.08152 • Published • 62
-
SELF: Language-Driven Self-Evolution for Large Language Model
Paper • 2310.00533 • Published • 2 -
GrowLength: Accelerating LLMs Pretraining by Progressively Growing Training Length
Paper • 2310.00576 • Published • 2 -
A Pretrainer's Guide to Training Data: Measuring the Effects of Data Age, Domain Coverage, Quality, & Toxicity
Paper • 2305.13169 • Published • 4 -
Transformers Can Achieve Length Generalization But Not Robustly
Paper • 2402.09371 • Published • 14
-
UniAudio: An Audio Foundation Model Toward Universal Audio Generation
Paper • 2310.00704 • Published • 20 -
Structural Similarities Between Language Models and Neural Response Measurements
Paper • 2306.01930 • Published • 2 -
Streaming Transformer ASR with Blockwise Synchronous Beam Search
Paper • 2006.14941 • Published • 2 -
NU-GAN: High resolution neural upsampling with GAN
Paper • 2010.11362 • Published • 2
-
JoMA: Demystifying Multilayer Transformers via JOint Dynamics of MLP and Attention
Paper • 2310.00535 • Published • 2 -
Solving Challenging Math Word Problems Using GPT-4 Code Interpreter with Code-based Self-Verification
Paper • 2308.07921 • Published • 24 -
AutoNumerics-Zero: Automated Discovery of State-of-the-Art Mathematical Functions
Paper • 2312.08472 • Published • 2 -
Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems
Paper • 2408.16293 • Published • 27
-
JoMA: Demystifying Multilayer Transformers via JOint Dynamics of MLP and Attention
Paper • 2310.00535 • Published • 2 -
Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small
Paper • 2211.00593 • Published • 2 -
Rethinking Interpretability in the Era of Large Language Models
Paper • 2402.01761 • Published • 23 -
Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla
Paper • 2307.09458 • Published • 12
-
A Biomedical Entity Extraction Pipeline for Oncology Health Records in Portuguese
Paper • 2304.08999 • Published • 3 -
CulturaX: A Cleaned, Enormous, and Multilingual Dataset for Large Language Models in 167 Languages
Paper • 2309.09400 • Published • 87 -
Robust Open-Vocabulary Translation from Visual Text Representations
Paper • 2104.08211 • Published • 1 -
Poro 34B and the Blessing of Multilinguality
Paper • 2404.01856 • Published • 15
-
MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data
Paper • 2304.08247 • Published • 2 -
Structural Similarities Between Language Models and Neural Response Measurements
Paper • 2306.01930 • Published • 2 -
Multimodal ChatGPT for Medical Applications: an Experimental Study of GPT-4V
Paper • 2310.19061 • Published • 8 -
Question-Answering Model for Schizophrenia Symptoms and Their Impact on Daily Life using Mental Health Forums Data
Paper • 2310.00448 • Published
-
Demystifying CLIP Data
Paper • 2309.16671 • Published • 20 -
Model Stock: All we need is just a few fine-tuned models
Paper • 2403.19522 • Published • 15 -
Bigger is not Always Better: Scaling Properties of Latent Diffusion Models
Paper • 2404.01367 • Published • 22 -
On the Scalability of Diffusion-based Text-to-Image Generation
Paper • 2404.02883 • Published • 19
-
Unified Functional Hashing in Automatic Machine Learning
Paper • 2302.05433 • Published • 2 -
Self-Discover: Large Language Models Self-Compose Reasoning Structures
Paper • 2402.03620 • Published • 117 -
Semi-Supervised Semantic Segmentation using Redesigned Self-Training for White Blood Cells
Paper • 2401.07278 • Published • 2
-
AutoNumerics-Zero: Automated Discovery of State-of-the-Art Mathematical Functions
Paper • 2312.08472 • Published • 2 -
MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
Paper • 2403.14624 • Published • 53 -
ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline
Paper • 2404.02893 • Published • 21 -
Rho-1: Not All Tokens Are What You Need
Paper • 2404.07965 • Published • 94
-
Unlocking the conversion of Web Screenshots into HTML Code with the WebSight Dataset
Paper • 2403.09029 • Published • 57 -
HuggingFaceM4/WebSight
Viewer • Updated • 2.75M • 14k • 393 -
HuggingFaceM4/VLM_WebSight_finetuned
Text Generation • 8B • Updated • 202 • 192 -
laion/filtered-wit
Viewer • Updated • 2.8M • 6.27k • 11
-
LocalMamba: Visual State Space Model with Windowed Selective Scan
Paper • 2403.09338 • Published • 8 -
SiMBA: Simplified Mamba-Based Architecture for Vision and Multivariate Time series
Paper • 2403.15360 • Published • 13 -
MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection
Paper • 2403.19888 • Published • 12
-
Functional Interpolation for Relative Positions Improves Long Context Transformers
Paper • 2310.04418 • Published • 4 -
SPBERT: An Efficient Pre-training BERT on SPARQL Queries for Question Answering over Knowledge Graphs
Paper • 2106.09997 • Published • 2 -
Neural Machine Translation of Rare Words with Subword Units
Paper • 1508.07909 • Published • 4 -
A Multimodal Approach to Device-Directed Speech Detection with Large Language Models
Paper • 2403.14438 • Published • 2
-
Video Mamba Suite: State Space Model as a Versatile Alternative for Video Understanding
Paper • 2403.09626 • Published • 15 -
VideoAgent: Long-form Video Understanding with Large Language Model as Agent
Paper • 2403.10517 • Published • 37 -
VSTAR: Generative Temporal Nursing for Longer Dynamic Video Synthesis
Paper • 2403.13501 • Published • 9 -
LITA: Language Instructed Temporal-Localization Assistant
Paper • 2403.19046 • Published • 19
-
Veagle: Advancements in Multimodal Representation Learning
Paper • 2403.08773 • Published • 10 -
mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality
Paper • 2304.14178 • Published • 3 -
Chart-based Reasoning: Transferring Capabilities from LLMs to VLMs
Paper • 2403.12596 • Published • 12 -
LLaVA-UHD: an LMM Perceiving Any Aspect Ratio and High-Resolution Images
Paper • 2403.11703 • Published • 17
-
Vision Transformer with Quadrangle Attention
Paper • 2303.15105 • Published • 2 -
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
Paper • 2103.14030 • Published • 5 -
MAFormer: A Transformer Network with Multi-scale Attention Fusion for Visual Recognition
Paper • 2209.01620 • Published • 2 -
CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows
Paper • 2107.00652 • Published • 2
-
StreamMultiDiffusion: Real-Time Interactive Generation with Region-Based Semantic Control
Paper • 2403.09055 • Published • 26 -
GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models
Paper • 2112.10741 • Published • 4 -
Lightweight Image Inpainting by Stripe Window Transformer with Joint Attention to CNN
Paper • 2301.00553 • Published • 3 -
ObjectDrop: Bootstrapping Counterfactuals for Photorealistic Object Removal and Insertion
Paper • 2403.18818 • Published • 28
-
StreamMultiDiffusion: Real-Time Interactive Generation with Region-Based Semantic Control
Paper • 2403.09055 • Published • 26 -
ReNoise: Real Image Inversion Through Iterative Noising
Paper • 2403.14602 • Published • 21 -
EdgeFusion: On-Device Text-to-Image Generation
Paper • 2404.11925 • Published • 23
-
Contrastive Decoding Improves Reasoning in Large Language Models
Paper • 2309.09117 • Published • 40 -
Chain-of-Thought Reasoning Without Prompting
Paper • 2402.10200 • Published • 110 -
MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
Paper • 2403.14624 • Published • 53 -
Chain of Thought Empowers Transformers to Solve Inherently Serial Problems
Paper • 2402.12875 • Published • 13
-
Exploring Large Language Models' Cognitive Moral Development through Defining Issues Test
Paper • 2309.13356 • Published • 38 -
Unveiling Safety Vulnerabilities of Large Language Models
Paper • 2311.04124 • Published • 9 -
TrustLLM: Trustworthiness in Large Language Models
Paper • 2401.05561 • Published • 69 -
Evaluating Frontier Models for Dangerous Capabilities
Paper • 2403.13793 • Published • 7
-
Table-GPT: Table-tuned GPT for Diverse Table Tasks
Paper • 2310.09263 • Published • 40 -
approximatelabs/tablib-v1-full
Viewer • Updated • 10.4B • 36 • 67 -
approximatelabs/tablib-v1-sample
Viewer • Updated • 44.9k • 109 • 14 -
TabLib: A Dataset of 627M Tables with Context
Paper • 2310.07875 • Published • 8
-
Qwen Technical Report
Paper • 2309.16609 • Published • 38 -
Qwen-VL: A Frontier Large Vision-Language Model with Versatile Abilities
Paper • 2308.12966 • Published • 12 -
Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models
Paper • 2311.07919 • Published • 9 -
Audio Dialogues: Dialogues dataset for audio and music understanding
Paper • 2404.07616 • Published • 15
significant improvements in zero-shot performance require exponentially more data, following a log-linear scaling trend
-
MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries
Paper • 2401.15391 • Published • 6 -
RAFT: Adapting Language Model to Domain Specific RAG
Paper • 2403.10131 • Published • 72 -
Superposition Prompting: Improving and Accelerating Retrieval-Augmented Generation
Paper • 2404.06910 • Published • 3 -
Stylus: Automatic Adapter Selection for Diffusion Models
Paper • 2404.18928 • Published • 15
-
MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries
Paper • 2401.15391 • Published • 6 -
Superposition Prompting: Improving and Accelerating Retrieval-Augmented Generation
Paper • 2404.06910 • Published • 3 -
Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models
Paper • 2404.18796 • Published • 71
-
CodeBERT: A Pre-Trained Model for Programming and Natural Languages
Paper • 2002.08155 • Published • 2 -
Text Generation with Diffusion Language Models: A Pre-training Approach with Continuous Paragraph Denoise
Paper • 2212.11685 • Published • 2 -
ReNoise: Real Image Inversion Through Iterative Noising
Paper • 2403.14602 • Published • 21 -
ByT5: Towards a token-free future with pre-trained byte-to-byte models
Paper • 2105.13626 • Published • 5
-
Streaming Transformer ASR with Blockwise Synchronous Beam Search
Paper • 2006.14941 • Published • 2 -
A Multimodal Approach to Device-Directed Speech Detection with Large Language Models
Paper • 2403.14438 • Published • 2 -
SONAR: Sentence-Level Multimodal and Language-Agnostic Representations
Paper • 2308.11466 • Published • 1
-
Neural networks behave as hash encoders: An empirical study
Paper • 2101.05490 • Published • 2 -
A Multiscale Visualization of Attention in the Transformer Model
Paper • 1906.05714 • Published • 2 -
BERT Rediscovers the Classical NLP Pipeline
Paper • 1905.05950 • Published • 3 -
Using Explainable AI and Transfer Learning to understand and predict the maintenance of Atlantic blocking with limited observational data
Paper • 2404.08613 • Published • 1
-
DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining
Paper • 2305.10429 • Published • 5 -
LLM2LLM: Boosting LLMs with Novel Iterative Data Enhancement
Paper • 2403.15042 • Published • 27 -
Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models
Paper • 2405.01535 • Published • 124 -
Discovering Preference Optimization Algorithms with and for Large Language Models
Paper • 2406.08414 • Published • 16
-
Adafactor: Adaptive Learning Rates with Sublinear Memory Cost
Paper • 1804.04235 • Published • 2 -
DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining
Paper • 2305.10429 • Published • 5 -
GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
Paper • 2305.13245 • Published • 6
-
DocVQA: A Dataset for VQA on Document Images
Paper • 2007.00398 • Published • 2 -
GQA: A New Dataset for Real-World Visual Reasoning and Compositional Question Answering
Paper • 1902.09506 • Published • 2 -
google/wit
Viewer • Updated • 2.66M • 468 • 68 -
Lin-Chen/MMStar
Viewer • Updated • 1.5k • 31k • 51
-
Neural Machine Translation of Rare Words with Subword Units
Paper • 1508.07909 • Published • 4 -
A Formal Perspective on Byte-Pair Encoding
Paper • 2306.16837 • Published • 3 -
Byte-Pair Encoding for Text-to-SQL Generation
Paper • 1910.08962 • Published • 2 -
Pattern Discovery in Time Series with Byte Pair Encoding
Paper • 2106.00614 • Published • 2
-
Recurrent Drafter for Fast Speculative Decoding in Large Language Models
Paper • 2403.09919 • Published • 21 -
SpecInfer: Accelerating Generative LLM Serving with Speculative Inference and Token Tree Verification
Paper • 2305.09781 • Published • 4 -
Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters
Paper • 2408.04093 • Published • 5
-
microsoft/table-transformer-detection
Object Detection • 28.8M • Updated • 3.08M • 421 -
RUCKBReasoning/TableLLM-13b
Text Generation • 13B • Updated • 108 • 32 -
RUCKBReasoning/TableLLM-7b
Text Generation • 7B • Updated • 57 • 16 -
TahaDouaji/detr-doc-table-detection
Object Detection • 41.6M • Updated • 234k • 63
-
Adding Conditional Control to Text-to-Image Diffusion Models
Paper • 2302.05543 • Published • 58 -
LightIt: Illumination Modeling and Control for Diffusion Models
Paper • 2403.10615 • Published • 18 -
SDXS: Real-Time One-Step Latent Diffusion Models with Image Conditions
Paper • 2403.16627 • Published • 22 -
DreamPolisher: Towards High-Quality Text-to-3D Generation via Geometric Diffusion
Paper • 2403.17237 • Published • 11
-
PERL: Parameter Efficient Reinforcement Learning from Human Feedback
Paper • 2403.10704 • Published • 60 -
WARM: On the Benefits of Weight Averaged Reward Models
Paper • 2401.12187 • Published • 19 -
RewardBench: Evaluating Reward Models for Language Modeling
Paper • 2403.13787 • Published • 22 -
DreamReward: Text-to-3D Generation with Human Preference
Paper • 2403.14613 • Published • 37
-
Vid2Robot: End-to-end Video-conditioned Policy Learning with Cross-Attention Transformers
Paper • 2403.12943 • Published • 15 -
Masked Audio Generation using a Single Non-Autoregressive Transformer
Paper • 2401.04577 • Published • 44 -
Cross-Attention Makes Inference Cumbersome in Text-to-Image Diffusion Models
Paper • 2404.02747 • Published • 13 -
InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation
Paper • 2404.02733 • Published • 22
-
LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression
Paper • 2403.12968 • Published • 26 -
Improving Text-to-Image Consistency via Automatic Prompt Optimization
Paper • 2403.17804 • Published • 19 -
The Unreasonable Effectiveness of Eccentric Automatic Prompts
Paper • 2402.10949 • Published • 5 -
Orca: Progressive Learning from Complex Explanation Traces of GPT-4
Paper • 2306.02707 • Published • 51
-
GaussianFlow: Splatting Gaussian Dynamics for 4D Content Creation
Paper • 2403.12365 • Published • 12 -
RadSplat: Radiance Field-Informed Gaussian Splatting for Robust Real-Time Rendering with 900+ FPS
Paper • 2403.13806 • Published • 19 -
Gaussian Frosting: Editable Complex Radiance Fields with Real-Time Rendering
Paper • 2403.14554 • Published • 14 -
DreamPolisher: Towards High-Quality Text-to-3D Generation via Geometric Diffusion
Paper • 2403.17237 • Published • 11
-
TexDreamer: Towards Zero-Shot High-Fidelity 3D Human Texture Generation
Paper • 2403.12906 • Published • 7 -
GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generation
Paper • 2403.14621 • Published • 16 -
LATTE3D: Large-scale Amortized Text-To-Enhanced3D Synthesis
Paper • 2403.15385 • Published • 8 -
LLaMA-Mesh: Unifying 3D Mesh Generation with Language Models
Paper • 2411.09595 • Published • 77
Converted the Elephants Never Forget paper to audio with Bark: https://drive.google.com/file/d/13IlbhKh71vxLpdYJ6mkIiiJZOUsf7XFv/view?usp=sharing
-
End-to-End Object Detection with Transformers
Paper • 2005.12872 • Published • 7 -
Elephants Never Forget: Memorization and Learning of Tabular Data in Large Language Models
Paper • 2404.06209 • Published • 5 -
TabReD: A Benchmark of Tabular Machine Learning in-the-Wild
Paper • 2406.19380 • Published • 49 -
SpreadsheetLLM: Encoding Spreadsheets for Large Language Models
Paper • 2407.09025 • Published • 140
-
End-to-End Object Detection with Transformers
Paper • 2005.12872 • Published • 7 -
COCONut: Modernizing COCO Segmentation
Paper • 2404.08639 • Published • 30 -
Grounded Language-Image Pre-training
Paper • 2112.03857 • Published • 3 -
Florence-2: Advancing a Unified Representation for a Variety of Vision Tasks
Paper • 2311.06242 • Published • 96
-
Evolutionary Optimization of Model Merging Recipes
Paper • 2403.13187 • Published • 58 -
Model Stock: All we need is just a few fine-tuned models
Paper • 2403.19522 • Published • 15 -
Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models
Paper • 2405.01535 • Published • 124
-
GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
Paper • 2305.13245 • Published • 6 -
Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models
Paper • 2404.12387 • Published • 40 -
OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework
Paper • 2404.14619 • Published • 126 -
Qwen2 Technical Report
Paper • 2407.10671 • Published • 172
-
MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
Paper • 2403.14624 • Published • 53 -
We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?
Paper • 2407.01284 • Published • 81 -
MAVIS: Mathematical Visual Instruction Tuning
Paper • 2407.08739 • Published • 32
-
MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
Paper • 2403.14624 • Published • 53 -
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?
Paper • 2407.01284 • Published • 81 -
DynaMath: A Dynamic Visual Benchmark for Evaluating Mathematical Reasoning Robustness of Vision Language Models
Paper • 2411.00836 • Published • 15
-
Explorative Inbetweening of Time and Space
Paper • 2403.14611 • Published • 13 -
StyleCineGAN: Landscape Cinemagraph Generation using a Pre-trained StyleGAN
Paper • 2403.14186 • Published • 10 -
Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction
Paper • 2403.18795 • Published • 20
-
Recourse for reclamation: Chatting with generative language models
Paper • 2403.14467 • Published • 8 -
Decoding Compressed Trust: Scrutinizing the Trustworthiness of Efficient LLMs Under Compression
Paper • 2403.15447 • Published • 16 -
Introducing v0.5 of the AI Safety Benchmark from MLCommons
Paper • 2404.12241 • Published • 13
-
Lexicon-Level Contrastive Visual-Grounding Improves Language Modeling
Paper • 2403.14551 • Published • 2 -
Adapting LLaMA Decoder to Vision Transformer
Paper • 2404.06773 • Published • 19 -
Toward a Better Understanding of Fourier Neural Operators: Analysis and Improvement from a Spectral Perspective
Paper • 2404.07200 • Published • 2 -
An inclusive review on deep learning techniques and their scope in handwriting recognition
Paper • 2404.08011 • Published • 1
-
LLM-Assisted Content Analysis: Using Large Language Models to Support Deductive Coding
Paper • 2306.14924 • Published • 2 -
When LLMs are Unfit Use FastFit: Fast and Effective Text Classification with Many Classes
Paper • 2404.12365 • Published • 2 -
In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering
Paper • 2311.06668 • Published • 5 -
Wave Network: An Ultra-Small Language Model
Paper • 2411.02674 • Published • 3
-
A Multimodal Approach to Device-Directed Speech Detection with Large Language Models
Paper • 2403.14438 • Published • 2 -
AniPortrait: Audio-Driven Synthesis of Photorealistic Portrait Animation
Paper • 2403.17694 • Published • 12 -
FlashSpeech: Efficient Zero-Shot Speech Synthesis
Paper • 2404.14700 • Published • 32 -
SONAR: Sentence-Level Multimodal and Language-Agnostic Representations
Paper • 2308.11466 • Published • 1
-
Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models
Paper • 2311.00871 • Published • 3 -
Can large language models explore in-context?
Paper • 2403.15371 • Published • 33 -
Data Distributional Properties Drive Emergent In-Context Learning in Transformers
Paper • 2205.05055 • Published • 2 -
Long-context LLMs Struggle with Long In-context Learning
Paper • 2404.02060 • Published • 37
Refer to additional papers: https://link.springer.com/article/10.1007/s10994-014-5458-8 and https://link.springer.com/article/10.1007/BF00992696
-
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
Paper • 2305.18290 • Published • 66 -
ICDPO: Effectively Borrowing Alignment Capability of Others via In-context Direct Preference Optimization
Paper • 2402.09320 • Published • 6 -
sDPO: Don't Use Your Data All at Once
Paper • 2403.19270 • Published • 41 -
Dueling RL: Reinforcement Learning with Trajectory Preferences
Paper • 2111.04850 • Published • 2
-
CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing
Paper • 2305.11738 • Published • 9 -
Shepherd: A Critic for Language Model Generation
Paper • 2308.04592 • Published • 33 -
CriticBench: Benchmarking LLMs for Critique-Correct Reasoning
Paper • 2402.14809 • Published • 3 -
DRLC: Reinforcement Learning with Dense Rewards from LLM Critic
Paper • 2401.07382 • Published • 2
-
Python Fuzzing for Trustworthy Machine Learning Frameworks
Paper • 2403.12723 • Published • 2 -
Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?
Paper • 2404.03411 • Published • 10 -
Teams of LLM Agents can Exploit Zero-Day Vulnerabilities
Paper • 2406.01637 • Published • 2 -
AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases
Paper • 2407.12784 • Published • 51
-
CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing
Paper • 2305.11738 • Published • 9 -
CriticBench: Benchmarking LLMs for Critique-Correct Reasoning
Paper • 2402.14809 • Published • 3 -
DRLC: Reinforcement Learning with Dense Rewards from LLM Critic
Paper • 2401.07382 • Published • 2
-
CriticBench: Benchmarking LLMs for Critique-Correct Reasoning
Paper • 2402.14809 • Published • 3 -
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
TACT: Advancing Complex Aggregative Reasoning with Information Extraction Tools
Paper • 2406.03618 • Published • 2
-
ReGAL: Refactoring Programs to Discover Generalizable Abstractions
Paper • 2401.16467 • Published • 10 -
Language Models as Compilers: Simulating Pseudocode Execution Improves Algorithmic Reasoning in Language Models
Paper • 2404.02575 • Published • 50 -
How Far Can We Go with Practical Function-Level Program Repair?
Paper • 2404.12833 • Published • 7
-
V-STaR: Training Verifiers for Self-Taught Reasoners
Paper • 2402.06457 • Published • 9 -
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
McEval: Massively Multilingual Code Evaluation
Paper • 2406.07436 • Published • 41 -
Is Programming by Example solved by LLMs?
Paper • 2406.08316 • Published • 13
-
V-STaR: Training Verifiers for Self-Taught Reasoners
Paper • 2402.06457 • Published • 9 -
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
Language Models as Compilers: Simulating Pseudocode Execution Improves Algorithmic Reasoning in Language Models
Paper • 2404.02575 • Published • 50
-
V-STaR: Training Verifiers for Self-Taught Reasoners
Paper • 2402.06457 • Published • 9 -
Self-Consistency Preference Optimization
Paper • 2411.04109 • Published • 19 -
Marco-o1: Towards Open Reasoning Models for Open-Ended Solutions
Paper • 2411.14405 • Published • 61 -
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Paper • 2402.03300 • Published • 145
-
ThemeStation: Generating Theme-Aware 3D Assets from Few Exemplars
Paper • 2403.15383 • Published • 15 -
FlexiDreamer: Single Image-to-3D Generation with FlexiCubes
Paper • 2404.00987 • Published • 23 -
MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs
Paper • 2402.15627 • Published • 36 -
Interactive3D: Create What You Want by Interactive 3D Generation
Paper • 2404.16510 • Published • 21
-
ThemeStation: Generating Theme-Aware 3D Assets from Few Exemplars
Paper • 2403.15383 • Published • 15 -
LATTE3D: Large-scale Amortized Text-To-Enhanced3D Synthesis
Paper • 2403.15385 • Published • 8 -
MaPa: Text-driven Photorealistic Material Painting for 3D Shapes
Paper • 2404.17569 • Published • 13 -
PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning
Paper • 2404.16994 • Published • 37
-
FollowIR: Evaluating and Teaching Information Retrieval Models to Follow Instructions
Paper • 2403.15246 • Published • 11 -
Noise-Aware Training of Layout-Aware Language Models
Paper • 2404.00488 • Published • 10 -
SnapKV: LLM Knows What You are Looking for Before Generation
Paper • 2404.14469 • Published • 27
-
FollowIR: Evaluating and Teaching Information Retrieval Models to Follow Instructions
Paper • 2403.15246 • Published • 11 -
Noise-Aware Training of Layout-Aware Language Models
Paper • 2404.00488 • Published • 10 -
Text Role Classification in Scientific Charts Using Multimodal Transformers
Paper • 2402.14579 • Published • 1
-
Compiler generated feedback for Large Language Models
Paper • 2403.14714 • Published • 7 -
Language Models as Compilers: Simulating Pseudocode Execution Improves Algorithmic Reasoning in Language Models
Paper • 2404.02575 • Published • 50 -
Compiling C to Safe Rust, Formalized
Paper • 2412.15042 • Published • 1
-
Beyond A*: Better Planning with Transformers via Search Dynamics Bootstrapping
Paper • 2402.14083 • Published • 47 -
GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints
Paper • 2305.13245 • Published • 6 -
Training a T5 Using Lab-sized Resources
Paper • 2208.12097 • Published • 1 -
Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints
Paper • 2212.05055 • Published • 6
-
EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters
Paper • 2402.04252 • Published • 30 -
MiniGPT4-Video: Advancing Multimodal LLMs for Video Understanding with Interleaved Visual-Textual Tokens
Paper • 2404.03413 • Published • 27 -
openai/clip-vit-large-patch14-336
Zero-Shot Image Classification • Updated • 9.33M • 307 -
openai/clip-vit-base-patch32
Zero-Shot Image Classification • Updated • 21.5M • 933
-
Prompt me a Dataset: An investigation of text-image prompting for historical image dataset creation using foundation models
Paper • 2309.01674 • Published • 2 -
Segment Anything
Paper • 2304.02643 • Published • 6 -
EgoLifter: Open-world 3D Segmentation for Egocentric Perception
Paper • 2403.18118 • Published • 12 -
A Multimodal Automated Interpretability Agent
Paper • 2404.14394 • Published • 23
-
Insightful analysis of historical sources at scales beyond human capabilities using unsupervised Machine Learning and XAI
Paper • 2310.09091 • Published • 2 -
Evolution and Transformation of Scientific Knowledge over the Sphaera Corpus: A Network Study
Paper • 2004.00520 • Published • 2 -
NAF-DPM: A Nonlinear Activation-Free Diffusion Probabilistic Model for Document Enhancement
Paper • 2404.05669 • Published • 1
-
InternLM2 Technical Report
Paper • 2403.17297 • Published • 34 -
sDPO: Don't Use Your Data All at Once
Paper • 2403.19270 • Published • 41 -
Learn Your Reference Model for Real Good Alignment
Paper • 2404.09656 • Published • 90 -
OpenBezoar: Small, Cost-Effective and Open Models Trained on Mixes of Instruction Data
Paper • 2404.12195 • Published • 12
-
LIMA: Less Is More for Alignment
Paper • 2305.11206 • Published • 27 -
Garment3DGen: 3D Garment Stylization and Texture Generation
Paper • 2403.18816 • Published • 25 -
EgoLifter: Open-world 3D Segmentation for Egocentric Perception
Paper • 2403.18118 • Published • 12 -
The Unreasonable Ineffectiveness of the Deeper Layers
Paper • 2403.17887 • Published • 83
Could also use a dna repo like: https://github.com/koeng101/dnadesign
-
One-step Diffusion with Distribution Matching Distillation
Paper • 2311.18828 • Published • 3 -
The Unreasonable Ineffectiveness of the Deeper Layers
Paper • 2403.17887 • Published • 83 -
Condition-Aware Neural Network for Controlled Image Generation
Paper • 2404.01143 • Published • 13 -
Locating and Editing Factual Associations in GPT
Paper • 2202.05262 • Published • 1
-
Lumiere: A Space-Time Diffusion Model for Video Generation
Paper • 2401.12945 • Published • 86 -
Long-form factuality in large language models
Paper • 2403.18802 • Published • 26 -
ObjectDrop: Bootstrapping Counterfactuals for Photorealistic Object Removal and Insertion
Paper • 2403.18818 • Published • 28 -
TC4D: Trajectory-Conditioned Text-to-4D Generation
Paper • 2403.17920 • Published • 18
-
The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
Paper • 1801.03924 • Published • 2 -
LLM2LLM: Boosting LLMs with Novel Iterative Data Enhancement
Paper • 2403.15042 • Published • 27 -
Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions
Paper • 1712.05884 • Published • 3 -
Efficient Memory Management for Large Language Model Serving with PagedAttention
Paper • 2309.06180 • Published • 54
-
The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
Paper • 1801.03924 • Published • 2 -
Fine-Tuning Language Models from Human Preferences
Paper • 1909.08593 • Published • 4 -
Training Verifiers to Solve Math Word Problems
Paper • 2110.14168 • Published • 7 -
Learning Transferable Visual Models From Natural Language Supervision
Paper • 2103.00020 • Published • 22
-
Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction
Paper • 2403.18795 • Published • 20 -
EgoLifter: Open-world 3D Segmentation for Egocentric Perception
Paper • 2403.18118 • Published • 12 -
GaussianCube: Structuring Gaussian Splatting using Optimal Transport for 3D Generative Modeling
Paper • 2403.19655 • Published • 19 -
Snap-it, Tap-it, Splat-it: Tactile-Informed 3D Gaussian Splatting for Reconstructing Challenging Surfaces
Paper • 2403.20275 • Published • 10
-
BioMedLM: A 2.7B Parameter Language Model Trained On Biomedical Text
Paper • 2403.18421 • Published • 23 -
Long-form factuality in large language models
Paper • 2403.18802 • Published • 26 -
stanford-crfm/BioMedLM
Text Generation • Updated • 5.78k • 454 -
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
Paper • 2305.18290 • Published • 66
-
Towards a World-English Language Model for On-Device Virtual Assistants
Paper • 2403.18783 • Published • 6 -
MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training
Paper • 2403.09611 • Published • 129 -
ReALM: Reference Resolution As Language Modeling
Paper • 2403.20329 • Published • 22 -
Ferret-UI: Grounded Mobile UI Understanding with Multimodal LLMs
Paper • 2404.05719 • Published • 83
-
Can large language models explore in-context?
Paper • 2403.15371 • Published • 33 -
GaussianCube: Structuring Gaussian Splatting using Optimal Transport for 3D Generative Modeling
Paper • 2403.19655 • Published • 19 -
WavLLM: Towards Robust and Adaptive Speech Large Language Model
Paper • 2404.00656 • Published • 11 -
Enabling Memory Safety of C Programs using LLMs
Paper • 2404.01096 • Published • 1
-
Can large language models explore in-context?
Paper • 2403.15371 • Published • 33 -
Long-context LLMs Struggle with Long In-context Learning
Paper • 2404.02060 • Published • 37 -
PIQA: Reasoning about Physical Commonsense in Natural Language
Paper • 1911.11641 • Published • 5 -
AQuA: A Benchmarking Tool for Label Quality Assessment
Paper • 2306.09467 • Published • 1
-
Generating Synthetic Computed Tomography for Radiotherapy: SynthRAD2023 Challenge Report
Paper • 2403.08447 • Published • 2 -
Deformable MRI Sequence Registration for AI-based Prostate Cancer Diagnosis
Paper • 2404.09666 • Published • 1 -
Surgical SAM 2: Real-time Segment Anything in Surgical Video by Efficient Frame Pruning
Paper • 2408.07931 • Published • 22
-
Mesh2NeRF: Direct Mesh Supervision for Neural Radiance Field Representation and Generation
Paper • 2403.19319 • Published • 14 -
Getting it Right: Improving Spatial Consistency in Text-to-Image Models
Paper • 2404.01197 • Published • 31 -
LLaVA-Gemma: Accelerating Multimodal Foundation Models with a Compact Language Model
Paper • 2404.01331 • Published • 27 -
LVLM-Intrepret: An Interpretability Tool for Large Vision-Language Models
Paper • 2404.03118 • Published • 25
-
TextCraftor: Your Text Encoder Can be Image Quality Controller
Paper • 2403.18978 • Published • 15 -
InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation
Paper • 2404.02733 • Published • 22 -
OmniFusion Technical Report
Paper • 2404.06212 • Published • 77 -
Transferable and Principled Efficiency for Open-Vocabulary Segmentation
Paper • 2404.07448 • Published • 12
-
LITA: Language Instructed Temporal-Localization Assistant
Paper • 2403.19046 • Published • 19 -
Snap-it, Tap-it, Splat-it: Tactile-Informed 3D Gaussian Splatting for Reconstructing Challenging Surfaces
Paper • 2403.20275 • Published • 10 -
Condition-Aware Neural Network for Controlled Image Generation
Paper • 2404.01143 • Published • 13 -
CantTalkAboutThis: Aligning Language Models to Stay on Topic in Dialogues
Paper • 2404.03820 • Published • 25
-
Dueling RL: Reinforcement Learning with Trajectory Preferences
Paper • 2111.04850 • Published • 2 -
Learning Trajectory Preferences for Manipulators via Iterative Improvement
Paper • 1306.6294 • Published • 3 -
Deep reinforcement learning from human preferences
Paper • 1706.03741 • Published • 4 -
Learning Dynamic Robot-to-Human Object Handover from Human Feedback
Paper • 1603.06390 • Published • 2
-
Learning Trajectory Preferences for Manipulators via Iterative Improvement
Paper • 1306.6294 • Published • 3 -
MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection
Paper • 2403.19888 • Published • 12 -
RL for Consistency Models: Faster Reward Guided Text-to-Image Generation
Paper • 2404.03673 • Published • 15 -
PhysDreamer: Physics-Based Interaction with 3D Objects via Video Generation
Paper • 2404.13026 • Published • 24
-
Fine-Tuning Language Models from Human Preferences
Paper • 1909.08593 • Published • 4 -
Transforming and Combining Rewards for Aligning Large Language Models
Paper • 2402.00742 • Published • 12 -
Leverage the Average: an Analysis of KL Regularization in RL
Paper • 2003.14089 • Published • 2 -
Direct Preference Optimization of Video Large Multimodal Models from Language Model Reward
Paper • 2404.01258 • Published • 12
https://web.stanford.edu/class/archive/stats/stats200/stats200.1172/Lecture24.pdf
-
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
Paper • 2305.18290 • Published • 66 -
HyperCLOVA X Technical Report
Paper • 2404.01954 • Published • 25 -
Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization
Paper • 2404.09956 • Published • 11 -
Learn Your Reference Model for Real Good Alignment
Paper • 2404.09656 • Published • 90
-
Transforming and Combining Rewards for Aligning Large Language Models
Paper • 2402.00742 • Published • 12 -
UltraFeedback: Boosting Language Models with High-quality Feedback
Paper • 2310.01377 • Published • 6 -
Learn Your Reference Model for Real Good Alignment
Paper • 2404.09656 • Published • 90 -
Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models
Paper • 2405.01535 • Published • 124
Figure 5: The probability mass assigned to partial human sentences. Flat distributions lead to many
moderately probable tokens, while peaked distribut
-
The Curious Case of Neural Text Degeneration
Paper • 1904.09751 • Published • 3 -
PIQA: Reasoning about Physical Commonsense in Natural Language
Paper • 1911.11641 • Published • 5 -
SocialIQA: Commonsense Reasoning about Social Interactions
Paper • 1904.09728 • Published • 4 -
HellaSwag: Can a Machine Really Finish Your Sentence?
Paper • 1905.07830 • Published • 8
-
The Curious Case of Neural Text Degeneration
Paper • 1904.09751 • Published • 3 -
Getting it Right: Improving Spatial Consistency in Text-to-Image Models
Paper • 2404.01197 • Published • 31 -
BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions
Paper • 1905.10044 • Published • 2 -
PIQA: Reasoning about Physical Commonsense in Natural Language
Paper • 1911.11641 • Published • 5
Spectrogram Prediction Network
As in Tacotron, mel spectrograms are computed through a short-
time Fourier transform (STFT) ... and a Hann window func
-
Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions
Paper • 1712.05884 • Published • 3 -
VoiceCraft: Zero-Shot Speech Editing and Text-to-Speech in the Wild
Paper • 2403.16973 • Published • 3 -
High Fidelity Neural Audio Compression
Paper • 2210.13438 • Published • 4 -
RALL-E: Robust Codec Language Modeling with Chain-of-Thought Prompting for Text-to-Speech Synthesis
Paper • 2404.03204 • Published • 9
-
Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions
Paper • 1712.05884 • Published • 3 -
Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization
Paper • 2404.09956 • Published • 11 -
Music Consistency Models
Paper • 2404.13358 • Published • 14 -
FlashSpeech: Efficient Zero-Shot Speech Synthesis
Paper • 2404.14700 • Published • 32
-
WhisperX: Time-Accurate Speech Transcription of Long-Form Audio
Paper • 2303.00747 • Published • 6 -
Custom Data Augmentation for low resource ASR using Bark and Retrieval-Based Voice Conversion
Paper • 2311.14836 • Published • 2 -
SONAR: Sentence-Level Multimodal and Language-Agnostic Representations
Paper • 2308.11466 • Published • 1 -
W2v-BERT: Combining Contrastive Learning and Masked Language Modeling for Self-Supervised Speech Pre-Training
Paper • 2108.06209 • Published • 1
-
keeeeenw/MicroLlama
Text Generation • 0.3B • Updated • 270 • 53 -
TinyLlama/TinyLlama-1.1B-intermediate-step-240k-503b
Text Generation • 1B • Updated • 1.49k • 20 -
TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
Text Generation • Updated • 105k • 188 -
TinyLlama/TinyLlama-1.1B-Chat-v1.0
Text Generation • 1B • Updated • 2.87M • • 1.58k
https://kyutai.org/
-
MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection
Paper • 2403.19888 • Published • 12 -
Measuring Style Similarity in Diffusion Models
Paper • 2404.01292 • Published • 16 -
GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding
Paper • 1804.07461 • Published • 4 -
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15
-
Gecko: Versatile Text Embeddings Distilled from Large Language Models
Paper • 2403.20327 • Published • 47 -
Round and Round We Go! What makes Rotary Positional Encodings useful?
Paper • 2410.06205 • Published • 2 -
Byte Latent Transformer: Patches Scale Better Than Tokens
Paper • 2412.09871 • Published • 108 -
MrT5: Dynamic Token Merging for Efficient Byte-level Language Models
Paper • 2410.20771 • Published • 3
-
Gecko: Versatile Text Embeddings Distilled from Large Language Models
Paper • 2403.20327 • Published • 47 -
2D Matryoshka Sentence Embeddings
Paper • 2402.14776 • Published • 8 -
Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference
Paper • 2412.13663 • Published • 163
Gradients flow differently for memorized and non-memorized during decoding
-
DiJiang: Efficient Large Language Models through Compact Kernelization
Paper • 2403.19928 • Published • 12 -
MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
Paper • 2309.12284 • Published • 19 -
TextHawk: Exploring Efficient Fine-Grained Perception of Multimodal Large Language Models
Paper • 2404.09204 • Published • 11 -
SAGS: Structure-Aware 3D Gaussian Splatting
Paper • 2404.19149 • Published • 14
-
The FinBen: An Holistic Financial Benchmark for Large Language Models
Paper • 2402.12659 • Published • 24 -
Long-context LLMs Struggle with Long In-context Learning
Paper • 2404.02060 • Published • 37 -
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding
Paper • 1804.07461 • Published • 4
-
DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
Paper • 1606.06160 • Published • 1 -
The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits
Paper • 2402.17764 • Published • 629 -
mobiuslabsgmbh/Llama-2-7b-chat-hf_1bitgs8_hqq
Text Generation • Updated • 25 • 74
-
Condition-Aware Neural Network for Controlled Image Generation
Paper • 2404.01143 • Published • 13 -
FlexiDreamer: Single Image-to-3D Generation with FlexiCubes
Paper • 2404.00987 • Published • 23 -
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline
Paper • 2404.02893 • Published • 21
Gradient-based surface extraction method
-
Non-Autoregressive Neural Machine Translation
Paper • 1711.02281 • Published • 1 -
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation
Paper • 2107.07651 • Published • 1 -
OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
Paper • 2404.07972 • Published • 52 -
AgentOhana: Design Unified Data and Training Pipeline for Effective Agent Learning
Paper • 2402.15506 • Published • 18
-
WavLLM: Towards Robust and Adaptive Speech Large Language Model
Paper • 2404.00656 • Published • 11 -
CameraCtrl: Enabling Camera Control for Text-to-Video Generation
Paper • 2404.02101 • Published • 24 -
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1
-
WavLLM: Towards Robust and Adaptive Speech Large Language Model
Paper • 2404.00656 • Published • 11 -
Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization
Paper • 2404.09956 • Published • 11 -
Long-form music generation with latent diffusion
Paper • 2404.10301 • Published • 27 -
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations
Paper • 2006.11477 • Published • 9
-
Noise-Aware Training of Layout-Aware Language Models
Paper • 2404.00488 • Published • 10 -
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking
Paper • 2204.08387 • Published • 8 -
LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding
Paper • 2012.14740 • Published • 3 -
LayoutLM: Pre-training of Text and Layout for Document Image Understanding
Paper • 1912.13318 • Published • 5
-
Noise-Aware Training of Layout-Aware Language Models
Paper • 2404.00488 • Published • 10 -
FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extraction
Paper • 2305.02549 • Published • 7 -
FormNet: Structural Encoding beyond Sequential Modeling in Form Document Information Extraction
Paper • 2203.08411 • Published • 1 -
ETC: Encoding Long and Structured Inputs in Transformers
Paper • 2004.08483 • Published • 1
-
Noise-Aware Training of Layout-Aware Language Models
Paper • 2404.00488 • Published • 10 -
FormNet: Structural Encoding beyond Sequential Modeling in Form Document Information Extraction
Paper • 2203.08411 • Published • 1 -
FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extraction
Paper • 2305.02549 • Published • 7 -
ETC: Encoding Long and Structured Inputs in Transformers
Paper • 2004.08483 • Published • 1
-
Streaming Dense Video Captioning
Paper • 2404.01297 • Published • 13 -
PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning
Paper • 2404.16994 • Published • 37 -
ShareGPT4Video: Improving Video Understanding and Generation with Better Captions
Paper • 2406.04325 • Published • 74
-
The state-of-the-art in Cardiac MRI Reconstruction: Results of the CMRxRecon Challenge in MICCAI 2023
Paper • 2404.01082 • Published • 1 -
Realism in Action: Anomaly-Aware Diagnosis of Brain Tumors from Medical Images Using YOLOv8 and DeiT
Paper • 2401.03302 • Published • 1 -
Brain2Music: Reconstructing Music from Human Brain Activity
Paper • 2307.11078 • Published • 42
-
FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extraction
Paper • 2305.02549 • Published • 7 -
FormNet: Structural Encoding beyond Sequential Modeling in Form Document Information Extraction
Paper • 2203.08411 • Published • 1 -
ETC: Encoding Long and Structured Inputs in Transformers
Paper • 2004.08483 • Published • 1 -
LongT5: Efficient Text-To-Text Transformer for Long Sequences
Paper • 2112.07916 • Published • 2
-
FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extraction
Paper • 2305.02549 • Published • 7 -
FormNet: Structural Encoding beyond Sequential Modeling in Form Document Information Extraction
Paper • 2203.08411 • Published • 1 -
More efficient manual review of automatically transcribed tabular data
Paper • 2306.16126 • Published • 1 -
CascadeTabNet: An approach for end to end table detection and structure recognition from image-based documents
Paper • 2004.12629 • Published • 3
https://github.com/learn2phoenix/CSD
-
CameraCtrl: Enabling Camera Control for Text-to-Video Generation
Paper • 2404.02101 • Published • 24 -
Adapting LLaMA Decoder to Vision Transformer
Paper • 2404.06773 • Published • 19 -
Interactive3D: Create What You Want by Interactive 3D Generation
Paper • 2404.16510 • Published • 21 -
Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
Paper • 2406.07394 • Published • 29
-
Long-context LLMs Struggle with Long In-context Learning
Paper • 2404.02060 • Published • 37 -
Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks
Paper • 2211.12588 • Published • 3 -
StructLM: Towards Building Generalist Models for Structured Knowledge Grounding
Paper • 2402.16671 • Published • 27 -
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model
Paper • 2404.04167 • Published • 13
-
Long-context LLMs Struggle with Long In-context Learning
Paper • 2404.02060 • Published • 37 -
Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks
Paper • 2211.12588 • Published • 3 -
StructLM: Towards Building Generalist Models for Structured Knowledge Grounding
Paper • 2402.16671 • Published • 27 -
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model
Paper • 2404.04167 • Published • 13
https://github.com/MoritzLaurer/zeroshot-classifier/tree/main
-
LLM-ABR: Designing Adaptive Bitrate Algorithms via Large Language Models
Paper • 2404.01617 • Published • 8 -
Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction
Paper • 2404.02905 • Published • 74 -
Learning From Mistakes Makes LLM Better Reasoner
Paper • 2310.20689 • Published • 29 -
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model
Paper • 2404.04167 • Published • 13
-
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline
Paper • 2404.02893 • Published • 21 -
MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
Paper • 2309.12284 • Published • 19 -
Premise Order Matters in Reasoning with Large Language Models
Paper • 2402.08939 • Published • 28
-
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
PointInfinity: Resolution-Invariant Point Diffusion Models
Paper • 2404.03566 • Published • 16 -
MonoPatchNeRF: Improving Neural Radiance Fields with Patch-based Monocular Guidance
Paper • 2404.08252 • Published • 6 -
SnapKV: LLM Knows What You are Looking for Before Generation
Paper • 2404.14469 • Published • 27
-
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 46 -
Locating and Editing Factual Associations in Mamba
Paper • 2404.03646 • Published • 3 -
Locating and Editing Factual Associations in GPT
Paper • 2202.05262 • Published • 1 -
KAN: Kolmogorov-Arnold Networks
Paper • 2404.19756 • Published • 116
-
LLaVA-Gemma: Accelerating Multimodal Foundation Models with a Compact Language Model
Paper • 2404.01331 • Published • 27 -
LVLM-Intrepret: An Interpretability Tool for Large Vision-Language Models
Paper • 2404.03118 • Published • 25 -
DesignQA: A Multimodal Benchmark for Evaluating Large Language Models' Understanding of Engineering Documentation
Paper • 2404.07917 • Published • 3 -
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32
-
LLaVA-Gemma: Accelerating Multimodal Foundation Models with a Compact Language Model
Paper • 2404.01331 • Published • 27 -
Data curation via joint example selection further accelerates multimodal learning
Paper • 2406.17711 • Published • 3 -
Unveiling Encoder-Free Vision-Language Models
Paper • 2406.11832 • Published • 55
-
LLaVA-Gemma: Accelerating Multimodal Foundation Models with a Compact Language Model
Paper • 2404.01331 • Published • 27 -
OmniFusion Technical Report
Paper • 2404.06212 • Published • 77 -
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15 -
WildGaussians: 3D Gaussian Splatting in the Wild
Paper • 2407.08447 • Published • 9
For HyperClova X they split 90% at 4096 and 10% at 32k context length during pt
Batches are grouped by similar token length to help optimize gpu/hardware. Mini batch lengths are different but the max number of tokens is the same.
-
HyperCLOVA X Technical Report
Paper • 2404.01954 • Published • 25 -
UltraFeedback: Boosting Language Models with High-quality Feedback
Paper • 2310.01377 • Published • 6 -
AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback
Paper • 2305.14387 • Published • 1 -
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Paper • 2402.03300 • Published • 145
-
On the Scalability of Diffusion-based Text-to-Image Generation
Paper • 2404.02883 • Published • 19 -
MonoPatchNeRF: Improving Neural Radiance Fields with Patch-based Monocular Guidance
Paper • 2404.08252 • Published • 6 -
Scaling Down to Scale Up: A Guide to Parameter-Efficient Fine-Tuning
Paper • 2303.15647 • Published • 4
-
On the Scalability of Diffusion-based Text-to-Image Generation
Paper • 2404.02883 • Published • 19 -
InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation
Paper • 2404.02733 • Published • 22 -
CoMat: Aligning Text-to-Image Diffusion Model with Image-to-Text Concept Matching
Paper • 2404.03653 • Published • 35 -
ControlNet++: Improving Conditional Controls with Efficient Consistency Feedback
Paper • 2404.07987 • Published • 48
-
Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction
Paper • 2404.02905 • Published • 74 -
ControlNet++: Improving Conditional Controls with Efficient Consistency Feedback
Paper • 2404.07987 • Published • 48 -
COCONut: Modernizing COCO Segmentation
Paper • 2404.08639 • Published • 30 -
MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs
Paper • 2402.15627 • Published • 36
-
Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction
Paper • 2404.02905 • Published • 74 -
On Speculative Decoding for Multimodal Large Language Models
Paper • 2404.08856 • Published • 13 -
Hydragen: High-Throughput LLM Inference with Shared Prefixes
Paper • 2402.05099 • Published • 20
-
Freditor: High-Fidelity and Transferable NeRF Editing by Frequency Decomposition
Paper • 2404.02514 • Published • 11 -
MonoPatchNeRF: Improving Neural Radiance Fields with Patch-based Monocular Guidance
Paper • 2404.08252 • Published • 6 -
Video2Game: Real-time, Interactive, Realistic and Browser-Compatible Environment from a Single Video
Paper • 2404.09833 • Published • 30 -
MeshLRM: Large Reconstruction Model for High-Quality Mesh
Paper • 2404.12385 • Published • 27
-
Freditor: High-Fidelity and Transferable NeRF Editing by Frequency Decomposition
Paper • 2404.02514 • Published • 11 -
BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer
Paper • 1904.06690 • Published • 1 -
Contrastive Chain-of-Thought Prompting
Paper • 2311.09277 • Published • 35 -
LLM-R2: A Large Language Model Enhanced Rule-based Rewrite System for Boosting Query Efficiency
Paper • 2404.12872 • Published • 11
-
Freditor: High-Fidelity and Transferable NeRF Editing by Frequency Decomposition
Paper • 2404.02514 • Published • 11 -
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model
Paper • 2404.04167 • Published • 13 -
Length Generalization of Causal Transformers without Position Encoding
Paper • 2404.12224 • Published • 1 -
Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
Paper • 2406.07394 • Published • 29
-
Event Camera Demosaicing via Swin Transformer and Pixel-focus Loss
Paper • 2404.02731 • Published • 1 -
MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
Paper • 2309.12284 • Published • 19 -
RALL-E: Robust Codec Language Modeling with Chain-of-Thought Prompting for Text-to-Speech Synthesis
Paper • 2404.03204 • Published • 9 -
Adapting LLaMA Decoder to Vision Transformer
Paper • 2404.06773 • Published • 19
-
I-Design: Personalized LLM Interior Designer
Paper • 2404.02838 • Published • 2 -
Scaling MLPs: A Tale of Inductive Bias
Paper • 2306.13575 • Published • 17 -
Fast Feedforward Networks
Paper • 2308.14711 • Published • 3 -
How Good Are Low-bit Quantized LLaMA3 Models? An Empirical Study
Paper • 2404.14047 • Published • 45
-
I am a Strange Dataset: Metalinguistic Tests for Language Models
Paper • 2401.05300 • Published • 3 -
Megalodon: Efficient LLM Pretraining and Inference with Unlimited Context Length
Paper • 2404.08801 • Published • 66 -
SpecInfer: Accelerating Generative LLM Serving with Speculative Inference and Token Tree Verification
Paper • 2305.09781 • Published • 4 -
MeshLRM: Large Reconstruction Model for High-Quality Mesh
Paper • 2404.12385 • Published • 27
-
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
Premise Order Matters in Reasoning with Large Language Models
Paper • 2402.08939 • Published • 28 -
Reasoning in Large Language Models: A Geometric Perspective
Paper • 2407.02678 • Published • 1 -
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Paper • 2402.03300 • Published • 145
-
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing
Paper • 2404.12253 • Published • 55 -
SEED-Bench-2-Plus: Benchmarking Multimodal Large Language Models with Text-Rich Visual Comprehension
Paper • 2404.16790 • Published • 10 -
A Thorough Examination of Decoding Methods in the Era of LLMs
Paper • 2402.06925 • Published • 1
-
Challenge LLMs to Reason About Reasoning: A Benchmark to Unveil Cognitive Depth in LLMs
Paper • 2312.17080 • Published • 1 -
Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
Paper • 2406.07394 • Published • 29 -
Qwen2 Technical Report
Paper • 2407.10671 • Published • 172 -
Self-Consistency Preference Optimization
Paper • 2411.04109 • Published • 19
Repo: https://github.com/HistLab/More-efficient-manual-review-of-automatically-transcribed-tabular-data
-
CascadeTabNet: An approach for end to end table detection and structure recognition from image-based documents
Paper • 2004.12629 • Published • 3 -
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking
Paper • 2204.08387 • Published • 8 -
Text Role Classification in Scientific Charts Using Multimodal Transformers
Paper • 2402.14579 • Published • 1 -
An inclusive review on deep learning techniques and their scope in handwriting recognition
Paper • 2404.08011 • Published • 1
-
LayoutLM: Pre-training of Text and Layout for Document Image Understanding
Paper • 1912.13318 • Published • 5 -
LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding
Paper • 2012.14740 • Published • 3 -
LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking
Paper • 2204.08387 • Published • 8
-
CodeEditorBench: Evaluating Code Editing Capability of Large Language Models
Paper • 2404.03543 • Published • 18 -
McEval: Massively Multilingual Code Evaluation
Paper • 2406.07436 • Published • 41 -
BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions
Paper • 2406.15877 • Published • 50 -
Qwen2 Technical Report
Paper • 2407.10671 • Published • 172
Table 5: Transformers struggle to learn Arithmetic Coding. In the sequence-to-sequence setting,
a model that learns AC compression/decompression shoul
Emergence with scale is unlikely Given the recent findings of [55], we anticipate that continuing
to scale models beyond 2 billion parameters is unlik
In this paper, we propose a strong alternative to PEFTs, LoReFT. LoReFT achieves strong per-
formance across benchmarks from four domains while being
-
SocialIQA: Commonsense Reasoning about Social Interactions
Paper • 1904.09728 • Published • 4 -
PIQA: Reasoning about Physical Commonsense in Natural Language
Paper • 1911.11641 • Published • 5 -
BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions
Paper • 1905.10044 • Published • 2 -
HellaSwag: Can a Machine Really Finish Your Sentence?
Paper • 1905.07830 • Published • 8
-
AQuA: A Benchmarking Tool for Label Quality Assessment
Paper • 2306.09467 • Published • 1 -
OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
Paper • 2404.07972 • Published • 52 -
BLINK: Multimodal Large Language Models Can See but Not Perceive
Paper • 2404.12390 • Published • 26 -
Vision language models are blind
Paper • 2407.06581 • Published • 84
MAWPS paper: https://aclanthology.org/N16-1136.pdf
-
Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problems
Paper • 1705.04146 • Published • 1 -
Training Verifiers to Solve Math Word Problems
Paper • 2110.14168 • Published • 7 -
Explaining Math Word Problem Solvers
Paper • 2307.13128 • Published • 1 -
MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms
Paper • 1905.13319 • Published • 2
https://github.com/google-deepmind/AQuA
-
Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problems
Paper • 1705.04146 • Published • 1 -
Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?
Paper • 2404.03411 • Published • 10 -
No "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance
Paper • 2404.04125 • Published • 29 -
Hydragen: High-Throughput LLM Inference with Shared Prefixes
Paper • 2402.05099 • Published • 20
Alpaca eval: https://github.com/tatsu-lab/alpaca_eval
-
UltraFeedback: Boosting Language Models with High-quality Feedback
Paper • 2310.01377 • Published • 6 -
Learn Your Reference Model for Real Good Alignment
Paper • 2404.09656 • Published • 90 -
Natural Language Reinforcement Learning
Paper • 2411.14251 • Published • 31 -
Group Robust Preference Optimization in Reward-free RLHF
Paper • 2405.20304 • Published • 1
-
RoBERTa: A Robustly Optimized BERT Pretraining Approach
Paper • 1907.11692 • Published • 10 -
Leveraging Pre-trained Checkpoints for Sequence Generation Tasks
Paper • 1907.12461 • Published • 1 -
Transformer Language Models without Positional Encodings Still Learn Positional Information
Paper • 2203.16634 • Published • 6 -
CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generation
Paper • 2102.04664 • Published • 2
-
MiniGPT4-Video: Advancing Multimodal LLMs for Video Understanding with Interleaved Visual-Textual Tokens
Paper • 2404.03413 • Published • 27 -
Scaling Data-Constrained Language Models
Paper • 2305.16264 • Published • 16 -
Emergence of Hidden Capabilities: Exploring Learning Dynamics in Concept Space
Paper • 2406.19370 • Published • 1
-
RALL-E: Robust Codec Language Modeling with Chain-of-Thought Prompting for Text-to-Speech Synthesis
Paper • 2404.03204 • Published • 9 -
Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models
Paper • 2311.07919 • Published • 9 -
FlashSpeech: Efficient Zero-Shot Speech Synthesis
Paper • 2404.14700 • Published • 32
-
Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?
Paper • 2404.03411 • Published • 10 -
The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions
Paper • 2404.13208 • Published • 40 -
A False Sense of Safety: Unsafe Information Leakage in 'Safe' AI Responses
Paper • 2407.02551 • Published • 9
-
CohereLabs/c4ai-command-r-plus
Text Generation • Updated • 2.81k • 1.78k -
CohereLabs/c4ai-command-r-plus-4bit
Text Generation • 105B • Updated • 37 • 258 -
mistral-community/Mixtral-8x22B-v0.1-4bit
Text Generation • 145B • Updated • 17 • 55 -
CohereLabs/c4ai-command-r-v01
Text Generation • Updated • 22.4k • 1.11k
-
openai/whisper-large-v3
Automatic Speech Recognition • 2B • Updated • 4.94M • • 5.68k -
openai/whisper-tiny
Automatic Speech Recognition • Updated • 768k • 427 -
openai/whisper-small
Automatic Speech Recognition • Updated • 2.28M • 558 -
openai/whisper-medium
Automatic Speech Recognition • Updated • 909k • 284
OpenAI has a 2024 tool referring to this technique: https://github.com/openai/transformer-debugger with https://transformer-circuits.pub/2023/monosema
Websit: https://rome.baulab.info/
-
Prompt-to-Prompt Image Editing with Cross Attention Control
Paper • 2208.01626 • Published • 3 -
BERT Rediscovers the Classical NLP Pipeline
Paper • 1905.05950 • Published • 3 -
A Multiscale Visualization of Attention in the Transformer Model
Paper • 1906.05714 • Published • 2 -
Analyzing Transformers in Embedding Space
Paper • 2209.02535 • Published • 3
-
Analyzing Transformers in Embedding Space
Paper • 2209.02535 • Published • 3 -
Prompt-to-Prompt Image Editing with Cross Attention Control
Paper • 2208.01626 • Published • 3 -
Dynamic Typography: Bringing Words to Life
Paper • 2404.11614 • Published • 46 -
Transformer Language Models without Positional Encodings Still Learn Positional Information
Paper • 2203.16634 • Published • 6
https://github.com/jessevig/bertviz
FFNs: Using a smoother GELU instead of an ReLu
-
DeViDe: Faceted medical knowledge for improved medical vision-language pre-training
Paper • 2404.03618 • Published • 2 -
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model
Paper • 2404.04167 • Published • 13 -
SpecInfer: Accelerating Generative LLM Serving with Speculative Inference and Token Tree Verification
Paper • 2305.09781 • Published • 4 -
McEval: Massively Multilingual Code Evaluation
Paper • 2406.07436 • Published • 41
Reward expressed as win-rates related to general preferences
A batched on-policy algorithm that conducts self-improvement iteratively via contrastive learning
a multi-step Markov Decision Process, allowing one to fine-tune consistency models toward a downstream task using just a reward function.
Focuses on policy improvement through search-based sampling
-
Realism in Action: Anomaly-Aware Diagnosis of Brain Tumors from Medical Images Using YOLOv8 and DeiT
Paper • 2401.03302 • Published • 1 -
MLP Can Be A Good Transformer Learner
Paper • 2404.05657 • Published • 1 -
Detecting and recognizing characters in Greek papyri with YOLOv8, DeiT and SimCLR
Paper • 2401.12513 • Published • 1 -
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets
Paper • 2404.02900 • Published • 1
-
MLP Can Be A Good Transformer Learner
Paper • 2404.05657 • Published • 1 -
Toward a Better Understanding of Fourier Neural Operators: Analysis and Improvement from a Spectral Perspective
Paper • 2404.07200 • Published • 2 -
An inclusive review on deep learning techniques and their scope in handwriting recognition
Paper • 2404.08011 • Published • 1 -
Long-form music generation with latent diffusion
Paper • 2404.10301 • Published • 27
-
Kandinsky: an Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion
Paper • 2310.03502 • Published • 79 -
Transferable and Principled Efficiency for Open-Vocabulary Segmentation
Paper • 2404.07448 • Published • 12 -
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32 -
COCONut: Modernizing COCO Segmentation
Paper • 2404.08639 • Published • 30
-
Kandinsky: an Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion
Paper • 2310.03502 • Published • 79 -
Transferable and Principled Efficiency for Open-Vocabulary Segmentation
Paper • 2404.07448 • Published • 12 -
RegionGPT: Towards Region Understanding Vision Language Model
Paper • 2403.02330 • Published • 2 -
GLIGEN: Open-Set Grounded Text-to-Image Generation
Paper • 2301.07093 • Published • 4
https://machinelearningmastery.com/how-to-implement-the-frechet-inception-distance-fid-from-scratch/
-
Kandinsky: an Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion
Paper • 2310.03502 • Published • 79 -
GLIGEN: Open-Set Grounded Text-to-Image Generation
Paper • 2301.07093 • Published • 4 -
Music Consistency Models
Paper • 2404.13358 • Published • 14 -
Align Your Steps: Optimizing Sampling Schedules in Diffusion Models
Paper • 2404.14507 • Published • 23
-
Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention
Paper • 2404.07143 • Published • 111 -
RULER: What's the Real Context Size of Your Long-Context Language Models?
Paper • 2404.06654 • Published • 41 -
An Evolved Universal Transformer Memory
Paper • 2410.13166 • Published • 6
-
DreamScene360: Unconstrained Text-to-3D Scene Generation with Panoramic Gaussian Splatting
Paper • 2404.06903 • Published • 21 -
CatLIP: CLIP-level Visual Recognition Accuracy with 2.7x Faster Pre-training on Web-scale Image-Text Data
Paper • 2404.15653 • Published • 29 -
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15 -
BlenderAlchemy: Editing 3D Graphics with Vision-Language Models
Paper • 2404.17672 • Published • 19
-
End-to-End Object Detection with Transformers
Paper • 2005.12872 • Published • 7 -
ConsistencyDet: Robust Object Detector with Denoising Paradigm of Consistency Model
Paper • 2404.07773 • Published • 1 -
Efficient Transformer Encoders for Mask2Former-style models
Paper • 2404.15244 • Published • 1 -
DETRs Beat YOLOs on Real-time Object Detection
Paper • 2304.08069 • Published • 16
-
ORPO: Monolithic Preference Optimization without Reference Model
Paper • 2403.07691 • Published • 72 -
ResearchAgent: Iterative Research Idea Generation over Scientific Literature with Large Language Models
Paper • 2404.07738 • Published • 2 -
Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models
Paper • 2405.01535 • Published • 124
Like coupled masses connected by springs
-
JetMoE: Reaching Llama2 Performance with 0.1M Dollars
Paper • 2404.07413 • Published • 38 -
Allowing humans to interactively guide machines where to look does not always improve a human-AI team's classification accuracy
Paper • 2404.05238 • Published • 3 -
Cognitive Architectures for Language Agents
Paper • 2309.02427 • Published • 8 -
Latent Positional Information is in the Self-Attention Variance of Transformer Language Models Without Positional Embeddings
Paper • 2305.13571 • Published • 2
Generalized multi head using RoPE
-
ConsistencyDet: Robust Object Detector with Denoising Paradigm of Consistency Model
Paper • 2404.07773 • Published • 1 -
Hyper-SD: Trajectory Segmented Consistency Model for Efficient Image Synthesis
Paper • 2404.13686 • Published • 29 -
Align Your Steps: Optimizing Sampling Schedules in Diffusion Models
Paper • 2404.14507 • Published • 23 -
NAF-DPM: A Nonlinear Activation-Free Diffusion Probabilistic Model for Document Enhancement
Paper • 2404.05669 • Published • 1
-
facebook/maskformer-swin-base-coco
Image Segmentation • 0.1B • Updated • 863 • • 26 -
facebook/mask2former-swin-small-coco-panoptic
Image Segmentation • 68.8M • Updated • 1.52k • 1 -
facebook/mask2former-swin-tiny-coco-instance
Image Segmentation • 47.5M • Updated • 57k • • 15 -
facebook/mask2former-swin-small-ade-semantic
Image Segmentation • 68.8M • Updated • 22.5k • • 8
Audio reading using bark: https://drive.google.com/file/d/1AlHLzeUd04LXgDj99SOvmQJTy9chufGo/view?usp=sharing
Read by Bark: https://drive.google.com/file/d/1apmyvLMEQ97ObHKzQna9URFHF0Xg-EsO/view?usp=sharing
-
ODA: Observation-Driven Agent for integrating LLMs and Knowledge Graphs
Paper • 2404.07677 • Published • 1 -
ResearchAgent: Iterative Research Idea Generation over Scientific Literature with Large Language Models
Paper • 2404.07738 • Published • 2 -
Scaling Instructable Agents Across Many Simulated Worlds
Paper • 2404.10179 • Published • 28 -
A Multimodal Automated Interpretability Agent
Paper • 2404.14394 • Published • 23
-
Rho-1: Not All Tokens Are What You Need
Paper • 2404.07965 • Published • 94 -
Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems
Paper • 2408.16293 • Published • 27 -
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Paper • 2402.03300 • Published • 145
-
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32 -
RegionGPT: Towards Region Understanding Vision Language Model
Paper • 2403.02330 • Published • 2 -
TextSquare: Scaling up Text-Centric Visual Instruction Tuning
Paper • 2404.12803 • Published • 30 -
Pegasus-v1 Technical Report
Paper • 2404.14687 • Published • 33
-
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32 -
GLIGEN: Open-Set Grounded Text-to-Image Generation
Paper • 2301.07093 • Published • 4 -
PhysDreamer: Physics-Based Interaction with 3D Objects via Video Generation
Paper • 2404.13026 • Published • 24 -
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15
Where the model is tasked with identifying
the object in a region mentioned in a query. we utilize the validation split of the
LVIS dataset
-
Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models
Paper • 2404.07973 • Published • 32 -
COCONut: Modernizing COCO Segmentation
Paper • 2404.08639 • Published • 30 -
GLIGEN: Open-Set Grounded Text-to-Image Generation
Paper • 2301.07093 • Published • 4 -
Grounded Language-Image Pre-training
Paper • 2112.03857 • Published • 3
Read by Bark: https://drive.google.com/file/d/1qltkY31-013JDQn-u2pmnjPyCaUcOqsV/view?usp=sharing
Read by Bark: https://drive.google.com/file/d/1no3kjSmexQxlS-KjhRB0jB5hz72Yuhsb/view?usp=sharing
Read by Bark: https://drive.google.com/file/d/1apmyvLMEQ97ObHKzQna9URFHF0Xg-EsO/view?usp=sharing
Read by Bark: https://drive.google.com/file/d/13IlbhKh71vxLpdYJ6mkIiiJZOUsf7XFv/view?usp=sharing
Read by Bark: https://drive.google.com/file/d/1AlHLzeUd04LXgDj99SOvmQJTy9chufGo/view?usp=sharing
-
mlabonne/AlphaMonarch-7B
Text Generation • 7B • Updated • 13.4k • • 148 -
Qwen/QwQ-32B-Preview
Text Generation • 33B • Updated • 10.3k • • 1.74k -
deepseek-ai/DeepSeek-R1-Distill-Llama-8B
Text Generation • 8B • Updated • 867k • • 859 -
deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
Text Generation • 8B • Updated • 567k • • 831
-
Instruction Tuning with Human Curriculum
Paper • 2310.09518 • Published • 3 -
A Thorough Examination of Decoding Methods in the Era of LLMs
Paper • 2402.06925 • Published • 1 -
Distilling System 2 into System 1
Paper • 2407.06023 • Published • 4 -
Byte Latent Transformer: Patches Scale Better Than Tokens
Paper • 2412.09871 • Published • 108
-
Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization
Paper • 2404.09956 • Published • 11 -
Contrastive Chain-of-Thought Prompting
Paper • 2311.09277 • Published • 35 -
FlashSpeech: Efficient Zero-Shot Speech Synthesis
Paper • 2404.14700 • Published • 32 -
AsyncDiff: Parallelizing Diffusion Models by Asynchronous Denoising
Paper • 2406.06911 • Published • 12
We use an ensemble filtering strategy based on two different CLAP models: 630k-audioset-best and 630k-best
-
ResearchAgent: Iterative Research Idea Generation over Scientific Literature with Large Language Models
Paper • 2404.07738 • Published • 2 -
Ag2Manip: Learning Novel Manipulation Skills with Agent-Agnostic Visual and Action Representations
Paper • 2404.17521 • Published • 13 -
LEGENT: Open Platform for Embodied Agents
Paper • 2404.18243 • Published • 22 -
Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
Paper • 2406.07394 • Published • 29
Reading by Bark: https://drive.google.com/file/d/1cdKmflJ3jRKszi5s3RF4Ru6nUw8tvpHX/view?usp=sharing reading duration 3 hours and 28 minutes
Read by Bark: https://drive.google.com/file/d/1yCc6rr199rQHHNwKozHhqzz0Rr48z03A/view?usp=sharing duration is 1 hour and 11 min 47s
Read by Bark: https://drive.google.com/file/d/1fEZ8uwnfniMljZ5S60NxOav6Qav2A-XB/view?usp=sharing duration is 44m 12s
better loss function. The sigmoid loss operates solely on image-text pairs and does not require a global view of the pairwise similarities
The DAC decoder tanh caused harmonic distortion
Found 5566 memorized, repeated audio sequences
an encoder-decoder model
which compresses videos to discrete embeddings (tokens) and a transformer model to translate
text embeddings to video tokens.
Fig 10. CFG substantially improves language conditionality.
The embeddings of
images and video patches from raw frames x are processed by a spatial and then a causal transformer
(AR in time) to gen video tokens
MaskGiT is trained to reconstruct
masked tokens z predicted by a frozen C-ViViT encoder and conditioned on T5X tokens of a given
prompt p0
-
On the Scalability of GNNs for Molecular Graphs
Paper • 2404.11568 • Published • 1 -
Octopus v4: Graph of language models
Paper • 2404.19296 • Published • 118 -
Architectures of Topological Deep Learning: A Survey on Topological Neural Networks
Paper • 2304.10031 • Published • 3 -
Meta Flow Matching: Integrating Vector Fields on the Wasserstein Manifold
Paper • 2408.14608 • Published • 8
1) random walk diagonals 2) Laplacian eigenvectors for geometry and position 3) global structural information about the graph
-
Explainable Lung Disease Classification from Chest X-Ray Images Utilizing Deep Learning and XAI
Paper • 2404.11428 • Published • 1 -
A Multimodal Automated Interpretability Agent
Paper • 2404.14394 • Published • 23 -
What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation
Paper • 2404.07129 • Published • 3 -
The Geometry of Categorical and Hierarchical Concepts in Large Language Models
Paper • 2406.01506 • Published • 3
Grad-CAM
LIME
-
MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs
Paper • 2402.15627 • Published • 36 -
One Wide Feedforward is All You Need
Paper • 2309.01826 • Published • 34 -
Fast Feedforward Networks
Paper • 2308.14711 • Published • 3 -
Memory Layers at Scale
Paper • 2412.09764 • Published • 5
See Page 7 Figure 5 on right: Repeating for 4 epochs is almost as good as new data
We propose to perform reward optimization using a RM trained for a different language. Assuming model generation quality transfers cross-lingually
contains documents and summaries in six languages (German, English, Spanish, Russian, Turkish, and Vietnamese) with pointwise human ratings
multilingual, pairwise human-rated chat transcripts.
For the SFT data, we use the human-preferred response in each pair to finetune the model
we recognize two memory bottlenecks: model weights and KV cache, and the latter gradually bottleneck(s) as context length increases
-
TriForce: Lossless Acceleration of Long Sequence Generation with Hierarchical Speculative Decoding
Paper • 2404.11912 • Published • 17 -
SnapKV: LLM Knows What You are Looking for Before Generation
Paper • 2404.14469 • Published • 27 -
LLM in a flash: Efficient Large Language Model Inference with Limited Memory
Paper • 2312.11514 • Published • 264 -
An Evolved Universal Transformer Memory
Paper • 2410.13166 • Published • 6
we utilize a 4K retrieval cache as an intermediate draft cache in our hierarchical system, while leveraging the JackFram/Llama68M (JF68M) [28] model
-
Dynamic Typography: Bringing Words to Life
Paper • 2404.11614 • Published • 46 -
Scene Coordinate Reconstruction: Posing of Image Collections via Incremental Learning of a Relocalizer
Paper • 2404.14351 • Published • 6 -
BlenderAlchemy: Editing 3D Graphics with Vision-Language Models
Paper • 2404.17672 • Published • 19 -
Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation
Paper • 2406.06525 • Published • 71
-
MeshLRM: Large Reconstruction Model for High-Quality Mesh
Paper • 2404.12385 • Published • 27 -
MaPa: Text-driven Photorealistic Material Painting for 3D Shapes
Paper • 2404.17569 • Published • 13 -
MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers
Paper • 2406.10163 • Published • 33 -
Meta 3D AssetGen: Text-to-Mesh Generation with High-Quality Geometry, Texture, and PBR Materials
Paper • 2407.02445 • Published • 4
Uniform scaling not as good as Head-based scaling
Page 7 fig shows NoPE extending passed the models context size from pretraining or fine-tuning
-
Length Generalization of Causal Transformers without Position Encoding
Paper • 2404.12224 • Published • 1 -
Transformer Language Models without Positional Encodings Still Learn Positional Information
Paper • 2203.16634 • Published • 6 -
Latent Positional Information is in the Self-Attention Variance of Transformer Language Models Without Positional Embeddings
Paper • 2305.13571 • Published • 2 -
The Impact of Positional Encoding on Length Generalization in Transformers
Paper • 2305.19466 • Published • 2
-
Latent Positional Information is in the Self-Attention Variance of Transformer Language Models Without Positional Embeddings
Paper • 2305.13571 • Published • 2 -
Transformers Can Represent n-gram Language Models
Paper • 2404.14994 • Published • 21 -
Are Sixteen Heads Really Better than One?
Paper • 1905.10650 • Published • 2 -
Reasoning in Large Language Models: A Geometric Perspective
Paper • 2407.02678 • Published • 1
-
Latent Positional Information is in the Self-Attention Variance of Transformer Language Models Without Positional Embeddings
Paper • 2305.13571 • Published • 2 -
BERTs are Generative In-Context Learners
Paper • 2406.04823 • Published • 1 -
Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference
Paper • 2412.13663 • Published • 163 -
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations
Paper • 2006.11477 • Published • 9
-
The Impact of Positional Encoding on Length Generalization in Transformers
Paper • 2305.19466 • Published • 2 -
Qwen2 Technical Report
Paper • 2407.10671 • Published • 172 -
Round and Round We Go! What makes Rotary Positional Encodings useful?
Paper • 2410.06205 • Published • 2 -
ThunderKittens: Simple, Fast, and Adorable AI Kernels
Paper • 2410.20399 • Published • 2
-
The Impact of Positional Encoding on Length Generalization in Transformers
Paper • 2305.19466 • Published • 2 -
Transformers Can Do Arithmetic with the Right Embeddings
Paper • 2405.17399 • Published • 54 -
Teaching Transformers Causal Reasoning through Axiomatic Training
Paper • 2407.07612 • Published • 2 -
Round and Round We Go! What makes Rotary Positional Encodings useful?
Paper • 2410.06205 • Published • 2
-
RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval
Paper • 2401.18059 • Published • 48 -
KGValidator: A Framework for Automatic Validation of Knowledge Graph Construction
Paper • 2404.15923 • Published • 2 -
The Geometry of Categorical and Hierarchical Concepts in Large Language Models
Paper • 2406.01506 • Published • 3
Core techniques: 1) unified grounding loss 2) language-aware deep fusion 3) pre-training with both types of data.
improved visual quality as the rough concept location and outline are decided in the early stages, followed by fine-grained details in later stages.
-
GLIGEN: Open-Set Grounded Text-to-Image Generation
Paper • 2301.07093 • Published • 4 -
YOLO-World: Real-Time Open-Vocabulary Object Detection
Paper • 2401.17270 • Published • 43 -
DETRs Beat YOLOs on Real-time Object Detection
Paper • 2304.08069 • Published • 16 -
RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformer
Paper • 2407.17140 • Published • 2
Driving particles
-
Hyper-SD: Trajectory Segmented Consistency Model for Efficient Image Synthesis
Paper • 2404.13686 • Published • 29 -
MultiBooth: Towards Generating All Your Concepts in an Image from Text
Paper • 2404.14239 • Published • 9 -
Stylus: Automatic Adapter Selection for Diffusion Models
Paper • 2404.18928 • Published • 15 -
MagicQuill: An Intelligent Interactive Image Editing System
Paper • 2411.09703 • Published • 80
-
How Good Are Low-bit Quantized LLaMA3 Models? An Empirical Study
Paper • 2404.14047 • Published • 45 -
Reasoning in Large Language Models: A Geometric Perspective
Paper • 2407.02678 • Published • 1 -
Natural Language Reinforcement Learning
Paper • 2411.14251 • Published • 31 -
Byte Latent Transformer: Patches Scale Better Than Tokens
Paper • 2412.09871 • Published • 108
-
How Good Are Low-bit Quantized LLaMA3 Models? An Empirical Study
Paper • 2404.14047 • Published • 45 -
LiteSearch: Efficacious Tree Search for LLM
Paper • 2407.00320 • Published • 40 -
Cut Your Losses in Large-Vocabulary Language Models
Paper • 2411.09009 • Published • 49 -
LLaMA-Mesh: Unifying 3D Mesh Generation with Language Models
Paper • 2411.09595 • Published • 77
Multi-modal Concept Extraction
-
All you need is a good init
Paper • 1511.06422 • Published • 1 -
Align Your Steps: Optimizing Sampling Schedules in Diffusion Models
Paper • 2404.14507 • Published • 23 -
Deep Residual Learning for Image Recognition
Paper • 1512.03385 • Published • 16 -
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15
-
All you need is a good init
Paper • 1511.06422 • Published • 1 -
Align Your Steps: Optimizing Sampling Schedules in Diffusion Models
Paper • 2404.14507 • Published • 23 -
Efficient Transformer Encoders for Mask2Former-style models
Paper • 2404.15244 • Published • 1 -
Deep Residual Learning for Image Recognition
Paper • 1512.03385 • Published • 16
-
OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework
Paper • 2404.14619 • Published • 126 -
Scaling Down to Scale Up: A Guide to Parameter-Efficient Fine-Tuning
Paper • 2303.15647 • Published • 4 -
Hyper-X: A Unified Hypernetwork for Multi-Task Multilingual Transfer
Paper • 2205.12148 • Published • 2 -
No More Adam: Learning Rate Scaling at Initialization is All You Need
Paper • 2412.11768 • Published • 43
-
OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework
Paper • 2404.14619 • Published • 126 -
apple/OpenELM-450M
Text Generation • 0.5B • Updated • 652 • 26 -
apple/OpenELM-3B
Text Generation • 3B • Updated • 401 • 130 -
apple/OpenELM-3B-Instruct
Text Generation • 3B • Updated • 3.24k • 339
Encoder - Weighted Stochastic Depth
https://stanfordnlp.github.io/CoreNLP/openie.html
See Appendix A.2
https://wordnet.princeton.edu/
Using SimCSE
-
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 15 -
Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation
Paper • 2406.06525 • Published • 71 -
Data curation via joint example selection further accelerates multimodal learning
Paper • 2406.17711 • Published • 3 -
Unveiling Encoder-Free Vision-Language Models
Paper • 2406.11832 • Published • 55
-
Interactive3D: Create What You Want by Interactive 3D Generation
Paper • 2404.16510 • Published • 21 -
SEED-Bench-2-Plus: Benchmarking Multimodal Large Language Models with Text-Rich Visual Comprehension
Paper • 2404.16790 • Published • 10 -
A Thorough Examination of Decoding Methods in the Era of LLMs
Paper • 2402.06925 • Published • 1 -
LLaVA-OneVision: Easy Visual Task Transfer
Paper • 2408.03326 • Published • 61
-
SEED-Bench-2-Plus: Benchmarking Multimodal Large Language Models with Text-Rich Visual Comprehension
Paper • 2404.16790 • Published • 10 -
MMWorld: Towards Multi-discipline Multi-faceted World Model Evaluation in Videos
Paper • 2406.08407 • Published • 28 -
GMAI-MMBench: A Comprehensive Multimodal Evaluation Benchmark Towards General Medical AI
Paper • 2408.03361 • Published • 85
-
LEGENT: Open Platform for Embodied Agents
Paper • 2404.18243 • Published • 22 -
Ag2Manip: Learning Novel Manipulation Skills with Agent-Agnostic Visual and Action Representations
Paper • 2404.17521 • Published • 13 -
Octopus v4: Graph of language models
Paper • 2404.19296 • Published • 118 -
AgentOhana: Design Unified Data and Training Pipeline for Effective Agent Learning
Paper • 2402.15506 • Published • 18
-
What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation
Paper • 2404.07129 • Published • 3 -
In-context Learning and Induction Heads
Paper • 2209.11895 • Published • 2 -
pyvene: A Library for Understanding and Improving PyTorch Models via Interventions
Paper • 2403.07809 • Published • 1
Two layer induction heads
-
What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation
Paper • 2404.07129 • Published • 3 -
In-context Learning and Induction Heads
Paper • 2209.11895 • Published • 2 -
pyvene: A Library for Understanding and Improving PyTorch Models via Interventions
Paper • 2403.07809 • Published • 1
See: pattern-preserving ablation
Additional reading: https://transformer-circuits.pub/2021/framework/index.html
-
In-context Learning and Induction Heads
Paper • 2209.11895 • Published • 2 -
What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation
Paper • 2404.07129 • Published • 3 -
pyvene: A Library for Understanding and Improving PyTorch Models via Interventions
Paper • 2403.07809 • Published • 1
Causal ablations taking into account LayerNorm
Modifying activations during training with proper gradient flow
See figure 6: Classes vs labels in columns B and C. Subcircuit B delays phase change on number classes vs C delays on number of labels (dramatically)
Larger vocab is better compression, but may result in longer training ICL phase change delays due to the slower Induction Head Copy Subcircuit (C)
See Appendix E: Merging Method Ablation on MoE Mistral and instruct 7B. Ties merged degenerate vs DARE model merges did not degenerate.
Local (gpu/cpu), compression (zstd) and over the wan (orc over s3 beats parquet too) results
See figure 19. Orc with cuDF beat parquet cuDF. Parquet arrow has dramatically more throughput without access to gpus
Additional paper with faq, code and tips on: https://github.com/microsoft/unilm/blob/master/bitnet/The-Era-of-1-bit-LLMs__Training_Tips_Code_FAQ.pdf
-
You Only Cache Once: Decoder-Decoder Architectures for Language Models
Paper • 2405.05254 • Published • 10 -
The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits
Paper • 2402.17764 • Published • 629 -
BitsFusion: 1.99 bits Weight Quantization of Diffusion Model
Paper • 2406.04333 • Published • 38 -
1-bit AI Infra: Part 1.1, Fast and Lossless BitNet b1.58 Inference on CPUs
Paper • 2410.16144 • Published • 5
-
Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
Paper • 2405.21060 • Published • 68 -
Chimera: Effectively Modeling Multivariate Time Series with 2-Dimensional State Space Models
Paper • 2406.04320 • Published • 10 -
MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection
Paper • 2403.19888 • Published • 12 -
Graph Mamba: Towards Learning on Graphs with State Space Models
Paper • 2402.08678 • Published • 17
Section 7.3.3
Appendix A.9
Table 7
Stage 2 training using LLaVA to describe the image with a caption
Guidance - "The intended effect is to decrease the diversity of the samples while increasing the quality of each individual sample."
-
Classifier-Free Diffusion Guidance
Paper • 2207.12598 • Published • 5 -
Adding Conditional Control to Text-to-Image Diffusion Models
Paper • 2302.05543 • Published • 58 -
Applying Guidance in a Limited Interval Improves Sample and Distribution Quality in Diffusion Models
Paper • 2404.07724 • Published • 14 -
Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation
Paper • 2406.06525 • Published • 71
-
Adding Conditional Control to Text-to-Image Diffusion Models
Paper • 2302.05543 • Published • 58 -
Smoothed Energy Guidance: Guiding Diffusion Models with Reduced Energy Curvature of Attention
Paper • 2408.00760 • Published • 7 -
MagicQuill: An Intelligent Interactive Image Editing System
Paper • 2411.09703 • Published • 80 -
BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusion
Paper • 2403.06976 • Published • 2
-
DataComp: In search of the next generation of multimodal datasets
Paper • 2304.14108 • Published • 2 -
The Synergy between Data and Multi-Modal Large Language Models: A Survey from Co-Development Perspective
Paper • 2407.08583 • Published • 13 -
TIP-I2V: A Million-Scale Real Text and Image Prompt Dataset for Image-to-Video Generation
Paper • 2411.04709 • Published • 27 -
YFCC100M: The New Data in Multimedia Research
Paper • 1503.01817 • Published • 1
-
QLoRA: Efficient Finetuning of Quantized LLMs
Paper • 2305.14314 • Published • 61 -
EfficientQAT: Efficient Quantization-Aware Training for Large Language Models
Paper • 2407.11062 • Published • 10 -
Spectra: A Comprehensive Study of Ternary, Quantized, and FP16 Language Models
Paper • 2407.12327 • Published • 79 -
BitNet a4.8: 4-bit Activations for 1-bit LLMs
Paper • 2411.04965 • Published • 69
-
CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs
Paper • 2406.18521 • Published • 31 -
We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?
Paper • 2407.01284 • Published • 81 -
ChartGemma: Visual Instruction-tuning for Chart Reasoning in the Wild
Paper • 2407.04172 • Published • 25
-
HuatuoGPT-Vision, Towards Injecting Medical Visual Knowledge into Multimodal LLMs at Scale
Paper • 2406.19280 • Published • 63 -
PPLLaVA: Varied Video Sequence Understanding With Prompt Guidance
Paper • 2411.02327 • Published • 11 -
MagicQuill: An Intelligent Interactive Image Editing System
Paper • 2411.09703 • Published • 80 -
LLaVA-o1: Let Vision Language Models Reason Step-by-Step
Paper • 2411.10440 • Published • 131
Repo: https://github.com/EleutherAI/sae
Deterministic methods with unaligned models usually perform better on all tasks except for open-ended text generation.
-
nomic-ai/nomic-embed-text-v1
Sentence Similarity • 0.1B • Updated • 7.33M • 567 -
nomic-ai/nomic-embed-text-v1.5-GGUF
Sentence Similarity • 0.1B • Updated • 90.3k • 103 -
BAAI/bge-large-en-v1.5
Feature Extraction • 0.3B • Updated • 15.4M • • 662 -
mixedbread-ai/mxbai-embed-2d-large-v1
Feature Extraction • 0.3B • Updated • 149k • • 43
Repo: https://github.com/shengliu66/ICV
-
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Efficient Algorithms for t-distributed Stochastic Neighborhood Embedding
Paper • 1712.09005 • Published • 1 -
UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
Paper • 1802.03426 • Published • 1 -
Principal subbundles for dimension reduction
Paper • 2307.03128 • Published • 1
-
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Barycentric Subspace Analysis on Manifolds
Paper • 1607.02833 • Published • 1 -
Template shape estimation: correcting an asymptotic bias
Paper • 1610.01502 • Published • 1 -
A Heat Diffusion Perspective on Geodesic Preserving Dimensionality Reduction
Paper • 2305.19043 • Published • 1
-
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Geodesic Multi-Modal Mixup for Robust Fine-Tuning
Paper • 2203.03897 • Published • 1 -
A Heat Diffusion Perspective on Geodesic Preserving Dimensionality Reduction
Paper • 2305.19043 • Published • 1 -
A micro Lie theory for state estimation in robotics
Paper • 1812.01537 • Published • 1
See also: https://dawn.cs.stanford.edu/2019/10/10/noneuclidean/
See also: https://dawn.cs.stanford.edu/2019/10/10/noneuclidean/
-
In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering
Paper • 2311.06668 • Published • 5 -
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Efficient Algorithms for t-distributed Stochastic Neighborhood Embedding
Paper • 1712.09005 • Published • 1
2008 paper: https://www.jmlr.org/papers/volume9/vandermaaten08a/vandermaaten08a.pdf
Recommended to explore constrained manifold areas with limited curvature (possible bias)
See also: https://dawn.cs.stanford.edu/2019/10/10/noneuclidean/
See also: "A Riemannian Framework for Tensor Computing" https://inria.hal.science/inria-00070743/file/RR-5255.pdf
-
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Barycentric Subspace Analysis on Manifolds
Paper • 1607.02833 • Published • 1 -
Poincaré Embeddings for Learning Hierarchical Representations
Paper • 1705.08039 • Published • 1 -
Efficient Algorithms for t-distributed Stochastic Neighborhood Embedding
Paper • 1712.09005 • Published • 1
Also see: https://dawn.cs.stanford.edu/2019/10/10/noneuclidean/
-
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
Paper • 2407.09468 • Published • 2 -
Architectures of Topological Deep Learning: A Survey on Topological Neural Networks
Paper • 2304.10031 • Published • 3 -
Galois Theory
Paper • 2408.07499 • Published • 1 -
Equivariant Transformer Networks
Paper • 1901.11399 • Published • 1
The manifold needs to be known for this layer to be implemented, and manifolds whose Exp enjoys an analytical expression are preferred
The manifold needs to be known for this layer to be implemented, and manifolds whose Log enjoys an analytical expression are preferred.
Euclidean signal on Euclidean domains for keys, queries and values, with group action on codomain.
Euclidean signal on manifold domain for all inputs / outputs, with domain group action.
Manifold-valued data convolutions. Tangent mean can be used for Fréchet mean to save on compute
-
Architectures of Topological Deep Learning: A Survey on Topological Neural Networks
Paper • 2304.10031 • Published • 3 -
Adaptive Topological Feature via Persistent Homology: Filtration Learning for Point Clouds
Paper • 2307.09259 • Published • 1 -
Persistent homology of the cosmic web. I: Hierarchical topology in ΛCDM cosmologies
Paper • 2011.12851 • Published • 1
-
Fixup Initialization: Residual Learning Without Normalization
Paper • 1901.09321 • Published • 1 -
RegMixup: Mixup as a Regularizer Can Surprisingly Improve Accuracy and Out Distribution Robustness
Paper • 2206.14502 • Published • 1 -
MixUp as Locally Linear Out-Of-Manifold Regularization
Paper • 1809.02499 • Published • 1
Repo: https://github.com/TorchJD/torchjd