Article
使用 DPO 微调 Llama 2
- +1
•
1
shipeng luo
luoagent
·
AI & ML interests
ML AI
Recent Activity
upvoted an article 3 days ago
使用 DPO 微调 Llama 2 upvoted a paper 4 days ago
On the Direction of RLVR Updates for LLM Reasoning: Identification and ExploitationOrganizations
None yet
upvoted an article 3 days ago
upvoted 8
papers 4 days ago
Astrolabe: Steering Forward-Process Reinforcement Learning for Distilled Autoregressive Video Models
Paper • 2603.17051 • Published • 106
On the Direction of RLVR Updates for LLM Reasoning: Identification and Exploitation
Paper • 2603.22117 • Published • 28
Sparse but Critical: A Token-Level Analysis of Distributional Shifts in RLVR Fine-Tuning of LLMs
Paper • 2603.22446 • Published • 7
CanViT: Toward Active-Vision Foundation Models
Paper • 2603.22570 • Published • 11
Rethinking Token-Level Policy Optimization for Multimodal Chain-of-Thought
Paper • 2603.22847 • Published • 25
CUA-Suite: Massive Human-annotated Video Demonstrations for Computer-Use Agents
Paper • 2603.24440 • Published • 91
MemMA: Coordinating the Memory Cycle through Multi-Agent Reasoning and In-Situ Self-Evolution
Paper • 2603.18718 • Published • 9
MuRF: Unlocking the Multi-Scale Potential of Vision Foundation Models
Paper • 2603.25744 • Published • 11
upvoted 11
papers 5 days ago
SpecEyes: Accelerating Agentic Multimodal LLMs via Speculative Perception and Planning
Paper • 2603.23483 • Published • 59
Qwen3-Coder-Next Technical Report
Paper • 2603.00729 • Published • 62
LongCat-Flash-Prover: Advancing Native Formal Reasoning via Agentic Tool-Integrated Reinforcement Learning
Paper • 2603.21065 • Published • 77
Intern-S1-Pro: Scientific Multimodal Foundation Model at Trillion Scale
Paper • 2603.25040 • Published • 123
Mixture-of-Depths Attention
Paper • 2603.15619 • Published • 79
Look Where It Matters: High-Resolution Crops Retrieval for Efficient VLMs
Paper • 2603.16932 • Published • 85
Penguin-VL: Exploring the Efficiency Limits of VLM with LLM-based Vision Encoders
Paper • 2603.06569 • Published • 117
OpenClaw-RL: Train Any Agent Simply by Talking
Paper • 2603.10165 • Published • 147
OpenSeeker: Democratizing Frontier Search Agents by Fully Open-Sourcing Training Data
Paper • 2603.15594 • Published • 148
Qianfan-OCR: A Unified End-to-End Model for Document Intelligence
Paper • 2603.13398 • Published • 152
OmniLottie: Generating Vector Animations via Parameterized Lottie Tokens
Paper • 2603.02138 • Published • 150