About

I am a master’s student in Data Science (DATS) at the University of Pennsylvania, advised by Jiatao Gu. I also work as a research intern at the VLAA Lab, UC Santa Cruz, supervised by Yuyin Zhou and Juncheng Wu. Before that, I received my bachelor’s degree in Computer Science from Tongji University.

Research Interests

VLM
Unified Multimodal ModelsMultimodal Reasoning
LLM
Latent Reasoning
Generative Models
Normalizing FlowsDiffusion Models

News

Show 5 earlierShow less
  • Apr 2026 Our paper Entropy-based Patchification Creates Semantic Tokens has been accepted by CVPR VisCon 2026 workshop!
  • Dec 2025 I win the Discover Award of Penn Wharton Innovation Fund!
  • Nov 2025 I join GMLR, the destination is star!
  • Aug 2025 I start my master in University of Pennsylvania! Let’s go quakers!
  • Jun 2025 We release MedFrameQA, a multiple image medical VQA benchmark to evaluate LLMs’ clinical reasoning capabilities.

Publications

  1. Preprint

    InteractionBench: A Real-Time Interaction Benchmark for Streaming Video Systems

    Enxin Song*, Suhao Yu*, Yifei Xu*, Barbara Su, Weili Xu, Wenhao Chai, Yao Tang, Jie Deng, Haiyang Xu, Jiatao Gu

  2. Preprint

    Learning from Teacher Continuations at Student States

    Haojin Wang*, Dylan Zhang*, Huaibo Chen, Suhao Yu, Yihang Sun, Zhanyang Jin, Jiaying Ye, Dianqi Li, Prasanna Sattigeri, Kamal Youcef-Toumi, Hao Peng

  3. COLM 2026Workshop on Efficient Reasoning

    Latent Reasoning with Normalizing Flows

    Guancheng Tu, Xiangjun Fu, Suhao Yu, Yao Tang, Haoqiang Kang, Lianhui Qin, Yizhe Zhang, Jiatao Gu

  4. Preprint

    Learning When to Think While Listening in Large Audio-Language Models

    Zhiyuan Song, Weici Zhao, Yang Xiao, Suhao Yu, Cheng Zhu, Jiatao Gu

  5. CVPR 2026VisCon Workshop

    Entropy-based Patchification Creates Semantic Tokens

    Suhao Yu, Jingjia Peng, Yao Tang, Jiatao Gu

  6. Preprint

    A Decade of Deep Learning-based Biomedical Image Segmentation

    Suhao Yu*, Haojin Wang*, Ningsen Wang*, Sicheng Chen*, Juncheng Wu, Zhenlong Yuan, Tianhao Qi, Zongwei Zhou, Fei Xia, Jun Ma, and Yuyin Zhou

  7. EMNLP 2026Main Conference

    MedFrameQA: A Multi-Image Medical VQA Benchmark for Clinical Reasoning

    Suhao Yu*, Haojin Wang*, Juncheng Wu*, Cihang Xie, Yuyin Zhou

* Equal contribution

Experience

  • FingerDance

    2024 – Present

    Deputy Technical DirectorAI-powered sign language translation startup

Service

  • Reviewer, ICLR 2027 2026
  • Reviewer, ACL Rolling Review (ARR) May 2026