单佳铭
单佳铭 Jiaming Shan

CS PhD of UCSB

I’m a CS PhD student in UCSB, advised by Prof. Yu Feng. My current research topic is Programming Language and Formal Verification. Now I’m working on projects that is related to

  • AI for math.
  • smart contracts

I’m also interested in AI, including training, embodied agents, etc.

Previously, I graduated from Shanghai Jiao Tong University ACM class, doing research about embodied AI and Foundation Models.


Publication

Building Cooperative Embodied Agents Modularly with Large Language Model

  • Authors: Hongxin Zhang*, Weihua Du*, Jiaming Shan, Qinhong Zhou, Yilun Du, Joshua B. Tenenbaum, Tianmin Shu, Chuang Gan.

  • ICLR, 2024 (poster)

  • We propose CoELA, a decentralized multi-agent system that uses GPT-4’s language abilities for effective cooperation in complex tasks. CoELA combines reasoning and communication skills with perception and memory, outperforming traditional methods. It shows enhanced cooperation with humans and potential for future multi-agent research.

Building Cooperative Embodied Agents Modularly with Large Language Model

  • Authors: Weihua Du, Qiushi Lyu, Jiaming Shan, Zhenting Qi, Hongxin Zhang, Sunli Chen, Andi Peng, Tianmin Shu, Kwonjoon Lee, Behzad Dariush, Chuang Gan

  • NeurIPS 2024 Track Datasets and Benchmarks Poster

  • We introduce Constrained Human-AI Cooperation (CHAIC), an embodied AI challenge centered on social intelligence. Agents must infer human intent and constraints and cooperatively assist with everyday tasks. Our benchmark features new constrained human agents and diverse, long-horizon indoor/outdoor scenarios. We evaluate baselines and propose a new LLM-based method, demonstrating CHAIC’s effectiveness in assessing machine social intelligence.

Download Résumé

Education

  1. Shanghai Jiao Tong University

    GPA (All Core Courses): 3.98/4.3, Ranking: 3/36

    GPA (2023 Courses): 4.11/4.3, Ranking: 1/35

Awards
National Scholarship Award
Shanghai Jiao Tong University ∙ October 2023