Zheyuan Brian Zhang

Robotics MS Student @ University of Michigan • zheyuan [at] umich [dot] edu


Ann Arbor, MI


My name is Zheyuan (Brian) Zhang (张喆元 in chinese). I am a master student at the University of Michigan pursuing Robotics MS in the Robotics Institute and Cognitive Science Graduate Certificate in the Weinberg Institute for Cognitive Science. Currently, I am a research assistant at the Situated Language and Embodied Dialogue (SLED) lab, advised by Professor Joyce Chai. Prior to UMich, I received my undergraduate degree in Computer Science (B.S.) from UMass Amherst, College of Information & Computer Sciences.

Click to view my latest Curriculum Vitae.

My research interests are centered at the nexus of artificial intelligence and cognitive science, driven by a desire to understand the intricacies of the brain and how can we learn so fast and so much from so little. Specifically, I am interested in the following topics:

  • Deep Learning: biologically-inspired neural architectures and learning algorithms (especially for aspects: “continual”, “online” and “few-shot”), compositionality
  • Natural Language Processing: human-like reasoning and planning, language grounding to vision and robotics, language acquisition and emergence
  • Robotics and Embodied Intelligence: manipulation, navigation, interactions (multi-agent interaction and human-robot interaction)
  • Cognitive Science: cognitive psychology, neuroscience


Jan 10, 2024 I am teaching EECS 492 - Introduction to Artificial Intelligence again as a Graduate Student Instructor (GSI) in Winter 2024 at UMich.
Oct 7, 2023 One paper accepted to EMNLP 2023 Main Conference.
Aug 29, 2023 I am teaching EECS 492 - Introduction to Artificial Intelligence as a Graduate Student Instructor (GSI) in Fall 2023 at UMich.

selected papers

  1. EMNLP
    From Heuristic to Analytic: Cognitively Motivated Strategies for Coherent Physical Commonsense Reasoning
    Zheyuan Zhang, Shane Storks, Fengyuan Hu, Sungryull Sohn, Moontae Lee, Honglak Lee, and Joyce Chai
    In Empirical Methods in Natural Language Processing (EMNLP) 2023
  2. arXiv
    Efficient In-Context Learning in Vision-Language Models for Egocentric Videos
    Keunwoo Peter Yu,  Zheyuan Zhang, Fengyuan Hu, and Joyce Chai
    In arXiv 2023
    Bot Lab: Autonomous Ground Vehicle from Low-level Control, SLAM to Planning and Exploration
    Zheyuan Zhang, Yu Zhu, Manu Aatitya Raajan Priyadharshini, and Thirumalaesh Ashokkumar
    In 2022
  4. INSAI
    Low-cost Solution for Vision-based Robotic Grasping
    Zheyuan Zhang, and Huiliang Shang
    In International Conference on Networking Systems of AI (IEEE INSAI) 2021