👋 About Me

Hello! My name is Xuyan Ye. I am a fourth-year undergraduate student majoring in AI at the Gaoling School of Artificial Intelligence, Renmin University of China, advised by Prof. Yankai Lin at RUCBM. I will join the Department of Computer Science and Engineering (CSE), Hong Kong University of Science and Technology (HKUST) as a Ph.D. student, advised by Prof. Song Guo. I am currently an LLM Research Intern at Ant Group, advised by Wei Wu, working on Agentic RL and LLM Reasoning.

I have also collaborated with Prof. Jiaxuan You and Prof. Heng Ji at UIUC. Previously, I spent a wonderful summer at BREATHE-AI, Nanyang Technological University, Singapore, conducting research under the guidance of Prof. Wei Lu.

My research interests include LLM Reasoning, Fine-Grained Agent Evaluation, and Process Supervision in Agentic RL.

🔥 News

📝 Preprints and Publications

NeurIPS 2026

The Many Faces of On-Policy Distillation: Pitfalls, Mechanisms, and Fixes

Siqi Zhu, Xuyan Ye^*, Hongyu Lu^*, Weiye Shi, Ge Liu^†

^* Equal contribution. ^† Corresponding author.

🌍 Homepage

📓 arXiv

🤗 Huggingface

💻 Github

Preprint

DARC: Decoupled Asymmetric Reasoning Curriculum for LLM Evolution

Shengda Fan^*, Xuyan Ye^*, Yankai Lin^†
* Equal contribution. ^† Corresponding author.

📓 arXiv

🤗 Huggingface

💻 Github

KDD D&B Track 2026

AgentProcessBench: Diagnosing Step-Level Process Quality in Tool-Using Agents

Shengda Fan^*, Xuyan Ye^*, Yupeng Huo, Zhiyuan Chen, Yiju Guo, Shenzhi Yang, Wenkai Yang, Shuqi Ye, Jingwen Chen, Haotian Chen, Xin Cong, Yankai Lin^†
* Equal contribution. ^† Corresponding author.

🌍 Homepage

📓 arXiv

🤗 Huggingface

💻 Github

Preprint

Co-Evolving Acting and Forecasting for Language Agents

Haofei Yu^*, Yixing Chen^*, Yining Zhao, Xuyan Ye, Yuxiang Liu, Zhigang Hua, Yan Xie, Shuang Yang, Jiaxuan You

^* Equal contribution.

📓 arXiv

💻 Github

Under Review

DeltaTTG: Training Multi-Turn Agents with a Token-to-Go Progress Meter

Xuyan Ye^*, Xiyan Fu^*, Shengda Fan^*, Jiawei Cui, Yankai Lin, Wei Lu^†

^* Equal contribution. ^† Corresponding author.

📓 arXiv

🤗 Huggingface

💻 Github

Preprint

Agent Error Dataset: Scaling 50,000 Error-Diagnosis Pairs for Failure Analysis and Error-Aware Post-Training

Kunlun Zhu, Xuyan Ye, Yibo Li, Cheng Qian, Beibin Li, Heng Ji^†

^† Corresponding author.

📓 arXiv

🤗 Huggingface

EMNLP Demo Track 2026

AgentDebugX: An Open-Source Toolkit for Failure Observability, Attribution, and Recovery in LLM Agents

Kunlun Zhu^*, Xuyan Ye^*, Zhiguang Han^*, Yuchen Zhao, Bingxuan Li, Weijia Zhang, Muxin Tian, Xiangru Tang, Pan Lu, James Zhou, Jiaxuan You, Heng Ji

^* Equal contribution. ^† Corresponding author.

🌍 Homepage

📓 arXiv

🤗 Huggingface

💻 Github

📖 Educations

• 2023.09 - 2027.06 (expected): Undergraduate student at the Gaoling School of Artificial Intelligence, Renmin University of China, Beijing.
• 2025.01 - 2025.03: Exchange student at the University of California, Davis, USA.

💻 Internships

• 2026.09 - Present: LLM Research Intern, Ant Group, Beijing.
• 2025.09 - 2025.11: LLM Application Intern, Content Quality and Data Service Platform, ByteDance.
• 2025.05 - 2025.08: AI Product Manager Intern, Large Language Model Division, Tencent.

🎖 Honors and Awards

• 2026.05 Beijing Natural Science Foundation Undergraduate “Qiyan” Project (启研计划), RMB 50,000.
• 2025.10 Renmin University of China Academic Excellence Scholarship, RMB 40,000.
• 2025.04 Renmin University of China Presidential Scholarship.

🙏 Acknowledgements

Beyond my heartfelt thanks to the professors mentioned above for their guidance and support, I would also like to thank the senior peers and friends who have helped me along the way.