10 papers
Do Prompt-Elicited Trajectories Reflect Training-Time Reward Hacking? A Systematic Study on Monitoring Training-Time Reward Hacking in Code Generation
Lichen Li, Hengguang Zhou, Yijun Liang +2
Reward hacking in code generation, where models exploit evaluation loopholes to obtain high reward without correctly solving the intended task, poses a critical challenge for Reinf…
Don't Think Longer, Think Wisely: Optimizing Thinking Dynamics for Large Reasoning Models
Sohyun An, Ruochen Wang, Tianyi Zhou +1
While recent success of large reasoning models (LRMs) significantly advanced LLMs' reasoning capability by optimizing the final answer accuracy using reinforcement learning, they m…
Exploring Expert Failures Improves LLM Agent Tuning
Li-Cheng Lan, Andrew Bai, Minhao Cheng +2
Large Language Models (LLMs) have shown tremendous potential as agents, excelling at tasks that require multiple rounds of reasoning and interactions. Rejection Sampling Fine-Tunin…
R1-Zero's "Aha Moment" in Visual Reasoning on a 2B Non-SFT Model
Hengguang Zhou, Xirui Li, Ruochen Wang +3
Recently DeepSeek R1 demonstrated how reinforcement learning with simple rule-based incentives can enable autonomous development of complex reasoning in large language models, char…
DrAttack: Prompt Decomposition and Reconstruction Makes Powerful LLM Jailbreakers
Xirui Li, Ruochen Wang, Minhao Cheng +2
The safety alignment of Large Language Models (LLMs) is vulnerable to both manual and automated jailbreak attacks, which adversarially trigger LLMs to output harmful content. Howev…
One Prompt is not Enough: Automated Construction of a Mixture-of-Expert Prompts
Ruochen Wang, Sohyun An, Minhao Cheng +3
Large Language Models (LLMs) exhibit strong generalization capabilities to novel tasks when prompted with language instructions and in-context demos. Since this ability sensitively…