Xincheng Tan

Xincheng Tan

M.S. student @ Zhejiang University

  • LLM / VLM Agents
  • Agentic RL
  • Neuro-Symbolic Planning

About me

I am a second-year master’s student in the School of Software Technology, Zhejiang University, working at the State Key Lab of CAD&CG. I am advised by Prof. Minfeng Zhu in the Visual Analytics and Intelligence Group. Before that, I received my bachelor’s degree in Computer Science and Technology from Southwest Jiaotong University in 2025.

My research interests include post-training and agentic RL for LLM / VLM agents, neuro-symbolic planning, and agent evaluation and failure attribution.

Publications

† Corresponding author.

Refuse and Recompute: What a Classical Planner Fixes in LLM Agents

Xincheng Tan, Ke Wang, Minfeng Zhu†

Manuscript, 2026

LLM agents that act over long horizons under partial observability lose mostly in two ways: an action after which the goal is unreachable, and acting without knowing what must still happen. A classical planner addresses both from the current state, with a guard that refuses such actions and by recomputing the remaining subgoals.

Education

  • M.S. in Software Engineering, Zhejiang University, 2025.09 – 2028.06 (expected)
    State Key Lab of CAD&CG. Average score: 91.4 / 100.
  • B.Eng. in Computer Science and Technology, Southwest Jiaotong University, 2021.09 – 2025.06
    GPA: 3.63 / 4.0; average score: 89.36 / 100; top 5% in major.

Honors

  • Five-Merit Graduate Student, Zhejiang University (2025–2026)
  • Outstanding Graduate of Sichuan Province (2025)
  • National Scholarship (2022–2023)
  • Merit Student Model, Southwest Jiaotong University (2021–2022)

Contact