Lizhu Zhang

AI Research Scientist Director at Meta, working on LLMs and RecSys

prof_pic.jpg

AI Researcher, husband, and dad of three. I head to the gym, hike, and play tennis to keep a fresh mind.

I am an AI Research Scientist Director at Meta. Before Meta, I worked on recommender systems and distributed systems at Twitter/X, Uber, Google, and VMware. My research spans reinforcement learning, causal inference, auto-research, and model paradigm innovation across RecSys and LLMs.

I feel fortunate to work alongside some of the brightest minds in the field, exploring the frontiers of AI together.

I received my bachelor’s degree in Physics and my master’s degree in Electrical Engineering from Tsinghua University, where I was advised by Dr. Yu Zheng.

news

May 15, 2026 Two papers accepted to ACL 2026Mixture-of-Minds: Multi-Agent Reinforcement Learning for Table Understanding (Main Conference) and TARo: Token-level Adaptive Routing for LLM Test-time Alignment (Findings).
May 01, 2026 Paper Token-Level LLM Collaboration via FusionRoute accepted to ICML 2026.
Mar 15, 2026 Paper StreamMem: Query-Agnostic KV Cache Memory for Streaming Video Understanding accepted to a CVPR 2026 Workshop (VidLLMs).

selected publications

* indicates corresponding / joint last authorship.  |  full list →

2026

  1. ICML
    Token-Level LLM Collaboration via FusionRoute
    Nuoya Xiong, Yuhang Zhou, Hanqing Zeng, Zhaorun Chen, Furong Huang, Shuchao Bi, Zhuokai Zhao, and Lizhu Zhang*
    In International Conference on Machine Learning (ICML), 2026
    Conference Accepted
  2. ACL
    Mixture-of-Minds: Multi-Agent Reinforcement Learning for Table Understanding
    Yuhang Zhou, Mingrui Zhang, Ke Li, Mingyi Wang, Qiao Liu, Qifei Wang, Jiayi Liu, Fei Liu, Serena Li, Weiwei Li, Mingze Gao, Abhishek Kumar, Xiangjun Fan, Zhuokai Zhao, and Lizhu Zhang*
    In Annual Meeting of the Association for Computational Linguistics (ACL), 2026
    Main Conference
  3. ACL
    TARo: Token-level Adaptive Routing for LLM Test-time Alignment
    Arushi Rai, Qiang Zhang, Hanqing Zeng, Yunkai Zhang, Dipesh Tamboli, Xiangjun Fan, Zhuokai Zhao, and Lizhu Zhang*
    In Annual Meeting of the Association for Computational Linguistics (ACL), 2026
    Findings
  4. ICML
    StreamMem: Query-Agnostic KV Cache Memory for Streaming Video Understanding
    Yanlai Yang, Zhuokai Zhao, Satya Narayan Shukla, Aashu Singh, Shlok Kumar Mishra, Lizhu Zhang*, and Mengye Ren
    In International Conference on Machine Learning (ICML), 2026
    Conference Submission

2025

  1. NeurIPS
    Thought Communication in Multiagent Collaboration
    Yujia Zheng, Zhuokai Zhao, Zijian Li, Yaqi Xie, Mingze Gao, Lizhu Zhang*, and Kun Zhang
    In Advances in Neural Information Processing Systems (NeurIPS), 2025
    Spotlight
  2. NeurIPS
    S’MoRE: Structural Mixture of Residual Experts for Parameter-Efficient LLM Fine-tuning
    Hanqing Zeng, Yinglong Xia, Zhuokai Zhao, Lizhu Zhang, Xiangjun Fan, and Benyu Zhang
    In Advances in Neural Information Processing Systems (NeurIPS), 2025
    Poster

2011

  1. TWEB
    Recommending friends and locations based on individual location history
    Yu Zheng, Lizhu Zhang, Zhengxin Ma, Xing Xie, and Wei-Ying Ma
    ACM Transactions on the Web, 2011

2009

  1. WWW
    Mining interesting locations and travel sequences from GPS trajectories
    Yu Zheng, Lizhu Zhang, Xing Xie, and Wei-Ying Ma
    In Proceedings of the 18th International Conference on World Wide Web (WWW), 2009
  2. GIS
    Mining correlation between locations using human location history
    Yu Zheng, Lizhu Zhang, Xing Xie, and Wei-Ying Ma
    In Proceedings of the 17th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (GIS), 2009