7 papers · 1 filter
EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments
Jundong Xu, Qingchuan Li, Jiaying Wu +11
Large language model (LLM) agents have achieved strong performance on a wide range of benchmarks, yet most evaluations assume static environments. In contrast, real-world deploymen…
TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning
Mingyue Cheng, Shuo Yu, Daoyu Wang +5
Spreadsheets and tables are widely used representations for structured data analysis, but effective analysis still requires substantial manual effort and domain expertise. Recent l…
AgentWorld: A Data Factory for Training LLM Agents from Multi-Turn Interactions
Daoyu Wang, Mingyue Cheng, Qingchuan Li +4
Training LLMs to act as capable agents requires learning from multi-turn interactions with environments. However, multi-turn interaction trajectories generated by heterogeneous age…
SocraticPO: Policy Optimization via Interactive Guidance
Zirui Liu, Tingyue Pan, Jie Ouyang +8
Reinforcement learning (RL) for large language models usually supervises reasoning with scalar outcome rewards, such as binary correctness. Such rewards provide an optimization dir…
CAPO: Critic-Guided Action-Aligned Policy Optimization for Advancing LLM Agent Capabilities
Daoyu Wang, Qingchuan Li, Mingyue Cheng +6
Reinforcement learning (RL) has become a key technique for improving the agentic capabilities of large language models (LLMs). Although critic-free methods such as GRPO are increas…
From Path Signatures to Sequential Modeling: Incremental Signature Contributions for Offline RL
Ziyi Zhao, Qingchuan Li, Yuxuan Xu
Path signatures embed trajectories into tensor algebra and constitute a universal, non-parametric representation of paths; however, in the standard form, they collapse temporal str…
PaperScout: An Autonomous Agent for Academic Paper Search with Process-Aware Sequence-Level Policy Optimization
Tingyue Pan, Jie Ouyang, Mingyue Cheng +7
Academic paper search is a fundamental task in scientific research, yet most existing approaches organize retrieval around predefined workflows or structured interaction protocols…