CV

Jiatong Zhao

zhaojiatong@sjtu.edu.cn
Shanghai, , CN

Summary

Undergraduate researcher working on reinforcement learning for reasoning, verifiable AI systems, and multimodal reasoning evaluation.

Education

  • Physics, Zhiyuan Honor College
    2027 expected
    Shanghai Jiao Tong University

Work Experience

  • Research Intern
    2025-09 -
    Working on scalable reasoning systems under the supervision of Ning Ding.
  • Research Intern
    2025-07 - 2025-12
    Worked on AI reasoning research under the supervision of Jie Fu.
  • Research Intern
    2025-03 - 2025-05
    Worked on GeoBench under the supervision of Junchi Yan and Renqiu Xia.
  • Research Assistant
    2024-08 - 2024-08
    Completed an AI4Physics research project.

Skills

Research Areas

  • Reinforcement learning for reasoning
  • Verifiable AI systems
  • Multimodal reasoning evaluation
  • AI4Physics

Publications

  • GeoBench: Rethinking Multimodal Geometric Problem-Solving via Hierarchical Evaluation
    2026
    ICLR 2026
    A hierarchical benchmark for multimodal geometry reasoning that decomposes failures into visual perception, goal-oriented planning, theorem application, and self-reflective backtracking.
  • Co4ICF: Co-evolving Physics-informed Surrogate and RL-based Pulse Optimizer for Inertial Confinement Fusion
    Manuscript
    A closed-loop AI4Physics system coupling a physics-informed surrogate with an RL-based optimizer for pulse design.

Interests

  • Scalable self-improving reasoning systems
    RL, verification, self-play, hierarchical evaluation