会员服务
Unlocking [CLS] Features for Continual Post-Training
Arxiv
0+阅读 · 2月19日
I-FailSense: Towards General Robotic Failure Detection with Vision-Language Models
I-FailSense: Towards General Robotic Failure Detection with Vision-Language Models
Arxiv
0+阅读 · 2月19日
NPG-Muse: Scaling Long Chain-of-Thought Reasoning with NP-Hard Graph Problems
Arxiv
0+阅读 · 2月17日
Improving Data and Reward Design for Scientific Reasoning in Large Language Models
Arxiv
0+阅读 · 2月9日
Improving Data and Reward Design for Scientific Reasoning in Large Language Models
Arxiv
0+阅读 · 2月10日
WoVR: World Models as Reliable Simulators for Post-Training VLA Policies with RL
Arxiv
0+阅读 · 2月15日
Less is Enough: Synthesizing Diverse Data in Feature Space of LLMs
Arxiv
0+阅读 · 2月12日
SIGMA: Selective-Interleaved Generation with Multi-Attribute Tokens
Arxiv
0+阅读 · 2月7日
On the optimization dynamics of RLVR: Gradient gap and step size thresholds
Arxiv
0+阅读 · 2月11日
Not All Layers Need Tuning: Selective Layer Restoration Recovers Diversity
Arxiv
0+阅读 · 2月6日
AWM: Accurate Weight-Matrix Fingerprint for Large Language Models
Arxiv
0+阅读 · 2月14日
TVCACHE: A Stateful Tool-Value Cache for Post-Training LLM Agents
Arxiv
0+阅读 · 2月11日
From SFT to RL: Demystifying the Post-Training Pipeline for LLM-based Vulnerability Detection
Arxiv
0+阅读 · 2月15日
ALOE: Action-Level Off-Policy Evaluation for Vision-Language-Action Model Post-Training
Arxiv
0+阅读 · 2月13日
ECHO-2: A Large-Scale Distributed Rollout Framework for Cost-Efficient Reinforcement Learning
Arxiv
0+阅读 · 2月10日
参考链接
微信扫码咨询专知VIP会员
Top