specimba's picture

In a Training Loop 🔄

specimba

specimba

·

AI & ML interests

everything

Recent Activity

new activity about 2 hours ago

specimba/nexus-os-v2-dualmode:Re: Final Followup

liked a model about 10 hours ago

deepseek-ai/DeepSeek-V4-Flash

updated a Space about 10 hours ago

build-small-hackathon/NEXUS-Visual-Weaver

View all activity

Organizations

upvoted a collection about 11 hours ago

Draft-OPD

6 items • Updated 5 days ago • 2

upvoted a paper about 11 hours ago

Draft-OPD: On-Policy Distillation for Speculative Draft Models

Paper • 2605.29343 • Published 13 days ago • 33

upvoted a paper about 14 hours ago

How to Evaluate Reward Models for RLHF

Paper • 2410.14872 • Published Oct 18, 2024 • 2

upvoted 4 papers 2 days ago

Unified-IO: A Unified Model for Vision, Language, and Multi-Modal Tasks

Paper • 2206.08916 • Published Jun 17, 2022 • 3

Do Androids Laugh at Electric Sheep? Humor "Understanding" Benchmarks from The New Yorker Caption Contest

Paper • 2209.06293 • Published Sep 13, 2022 • 1

MAUVE Scores for Generative Models: Theory and Practice

Paper • 2212.14578 • Published Dec 30, 2022 • 1

HellaSwag: Can a Machine Really Finish Your Sentence?

Paper • 1905.07830 • Published May 19, 2019 • 9

upvoted a collection 3 days ago

Kontext Dev LoRAs

Collection of Kontext Dev LoRAs by fal • 30 items • Updated Jul 27, 2025 • 37

upvoted 3 papers 4 days ago

AVControl: Efficient Framework for Training Audio-Visual Controls

Paper • 2603.24793 • Published Mar 25 • 29

HDR Video Generation via Latent Alignment with Logarithmic Encoding

Paper • 2604.11788 • Published Apr 13 • 13

LTX-2: Efficient Joint Audio-Visual Foundation Model

Paper • 2601.03233 • Published Jan 6 • 181

upvoted a collection 5 days ago

JOSIE

The JOSIE models are a family of uncensored, high-performance language models developed by Gökdeniz Gülmez. • 15 items • Updated Mar 9 • 2

upvoted 5 papers 5 days ago

Gabliteration: Adaptive Multi-Directional Neural Weight Modification for Selective Behavioral Alteration in Large Language Models

Paper • 2512.18901 • Published Dec 21, 2025 • 5

Future-KL Regularized GRPO: Process-Level Credit Assignment from f-Divergence Regularization

Paper • 2601.10201 • Published 18 days ago • 10

CLEANER: Self-Purified Trajectories Boost Agentic Reinforcement Learning

Paper • 2601.15141 • Published Jan 21 • 3

Self-Distilled Reasoner: On-Policy Self-Distillation for Large Language Models

Paper • 2601.18734 • Published Jan 26 • 8

Reinforcement Learning via Self-Distillation

Paper • 2601.20802 • Published Jan 28 • 49

upvoted a changelog 8 days ago

Hugging Face Changelog

Filter Models page by Base Models only

12 days ago

• 144

upvoted a paper 8 days ago

Mellum2 Technical Report

Paper • 2605.31268 • Published 12 days ago • 54

upvoted a collection 8 days ago

Mellum 2

Mellum2 model weights • 6 items • Updated 8 days ago • 107