Situo Zhang

Situo Zhang 张思拓

I am a fourth-year Ph.D. student @ Computer Science at Shanghai Jiao Tong University (SJTU). I am with the X-LANCE Lab, supervised by Professor Kai Yu and Associate Professor Lu Chen. I am currently a research intern at Shanghai AI Laboratory. Previously, I received my B.Sc. degree in Computer Science and Technology from Nanjing University (NJU) in 2022.

My research interests lie in large language model reasoning and agentic reinforcement learning, with a particular focus on GUI agents and autonomous AI scientists. My long-term goal is to build agents that can reason, learn from environmental feedback, and reliably solve complex real-world and scientific problems.

Photo @ Qiandao Lake, China · 2026

News

Publications

* Equal contribution

CharTool motivation and tool-integrated chart reasoning overview

ACM MM 2026

CharTool: Tool-Integrated Visual Reasoning for Chart Understanding

Situo Zhang*, Yifan Zhang*, Zichen Zhu, Da Ma, Lei Pan, Danyang Zhang, Zihan Zhao, Lu Chen, Kai Yu

We equip multimodal LLMs with image cropping and code-based computation, then train tool use through agentic reinforcement learning on DuoChart. This improves fine-grained visual grounding and numerical reasoning across chart benchmarks.

Visual ReasoningTool Use
Comparison of imitation learning, outcome rewards, and progress rewards

arXiv 2025

ProgRM: Build Better GUI Agents with Progress Rewards

Danyang Zhang*, Situo Zhang*, Ziyue Yang, Zichen Zhu, Zihan Zhao, Ruisheng Cao, Lu Chen, Kai Yu

We provide dense, step-level progress rewards for training GUI agents instead of scoring only final outcomes. An LCS-based self-annotation method identifies key trajectory steps and supplies progress labels without costly manual annotation.

GUI AgentsReward Modeling
Pacer dynamic-window speculative decoding overview

IEEE/ACM TASLP

Pacer: Blockwise Pre-verification for Speculative Decoding with Adaptive Length

Situo Zhang, Yifan Zhang, Zichen Zhu, Hankun Wang, Da Ma, Danyang Zhang, Lu Chen, Kai Yu

We dynamically adjust speculative draft length through lightweight blockwise pre-verification, stopping low-quality drafts before target-model verification. Pacer consistently improves standard speculative decoding and achieves up to 2.66× speedup over autoregressive decoding.

Efficient InferenceSpeculative Decoding
Additional publications +

Collaboration

I collaborate closely with Danyang Zhang, Ruisheng Cao @ Qwen, Alibaba, Zihan Zhao, Zichen Zhu, Hongshen Xu @ MiMo, Xiaomi, Da Ma, Yifan Zhang, Xuanze Lin, and Hanqi Li. I welcome opportunities for discussions and potential collaborations. Please feel free to contact me via email.

Academic Service

Reviewer for Conferences

  • 2025, 2026
    International Conference on Machine Learning (ICML)
  • 2025
    Conference on Neural Information Processing Systems (NeurIPS)
  • 2024, 2026
    Conference on Empirical Methods in Natural Language Processing (EMNLP)
  • 2025
    Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL)
  • 2026
    Conference on Language Modeling (COLM)

Reviewer for Journals

  • Transactions on Machine Learning Research (TMLR)
  • Nature (Co-reviewer)
  • Nature Communications (Co-reviewer)

Teaching

Teaching Assistant, Natural Language Processing, Fall 2023
Teaching Assistant, Discrete Mathematics, Spring 2024

Selected Honors