11 papers
Skill-Use: Can LLMs Actually Use Skills in Agentic Harnesses?
Jinyi Han, Yuanjian Xu, Ying Liao +6
Large language model (LLM) agents increasingly rely on skills, structured documents that specify when to act, which procedure to follow, and which tools are allowed. Existing evalu…
From Outcomes to Actions: Leveraging Hindsight for Long-Horizon Language Agent Training
Zishang Jiang, Tingyun Li, Jinyi Han +7
Reinforcement learning (RL) has become a widely adopted technique for improving large language models (LLMs) on complex tasks. Despite this progress, existing RL methods still face…
ADaPT: Token-Level Decoupling for Efficient Large Reasoning Models
Tingyun Li, Zishang Jiang, Jinyi Han +8
Large reasoning models rely on long chain-of-thought to achieve strong performance, but applying such reasoning uniformly incurs high computational cost. Existing efficiency-orient…
Learning from Own Solutions: Self-Conditioned Credit Assignment for Reinforcement Learning with Verifiable Rewards
Yingyu Shan, Yuhang Guo, Zihao Cheng +7
Reinforcement learning with verifiable rewards (RLVR) has driven substantial progress in training LLMs for reasoning tasks, but representative methods such as GRPO assign uniform c…
DiffCVaR: Reinforcement Learning for Risk Adaptation via Differentiable CVaR Barrier Functions
Xinyi Wang, Taekyung Kim, Bardh Hoxha +2
Planning through crowded environments under uncertain obstacle motions remains difficult, as stochastic interactions often induce overly conservative behavior or reduced efficiency…
Terminal-World: Scaling Terminal-Agent Environments via Agent Skills
Zihao Cheng, Hongru Wang, Zeming Liu +6
Terminal agents extend Large Language Models with the ability to execute tasks directly in command-line environments, but their progress is bottlenecked by the scarcity of high-qua…
Don't Tell the Answer, Truly Guide the Reasoning During RL Rollouts
Xinyi Wang, Jinyi Han, Zishang Jiang +7
Reinforcement Learning (RL) has become a key driver for enhancing the long chain-of-thought (CoT) reasoning capabilities of Large Language Models (LLMs). However, prevalent methods…
Selective Expert Guidance for Effective and Diverse Exploration in Reinforcement Learning of LLMs
Zishang Jiang, Jinyi Han, Tingyun Li +7
Reinforcement Learning with Verifiable Rewards (RLVR) has become a widely adopted technique for enhancing the reasoning ability of Large Language Models (LLMs). However, the effect…
Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking
Jinyi Han, Ying Huang, Ying Liao +11
Large Reasoning Models (LRMs) have achieved impressive performance on challenging tasks, yet their deep reasoning often incurs substantial computational costs. To achieve efficient…
Mind the Generation Process: Fine-Grained Confidence Estimation During LLM Generation
Jinyi Han, Tingyun Li, Shisong Chen +8
While large language models (LLMs) have demonstrated remarkable performance across diverse tasks, they fundamentally lack self-awareness and frequently exhibit overconfidence, assi…
A Stitch in Time Saves Nine: Proactive Self-Refinement for Language Models
Jinyi Han, Xinyi Wang, Haiquan Zhao +9
Recent advances in self-refinement have demonstrated significant potential for improving the outputs of large language models (LLMs) through iterative refinement. However, most exi…