Longtao Zheng 郑龙韬

Longtao Zheng is a PhD student at the College of Computing and Data Science, Nanyang Technological University (NTU) Singapore, advised by Prof. Bo An. Previously, he obtained his Bachelor's degree in computer science from University of Science and Technology of China (USTC) in 2022. His research focuses on foundation models for decision making, controllable video generation, and reinforcement learning.

News
  • Check out our latest work on general computer agents: AgentStudio and Cradle!
  • One paper is accepted by KDD 2024 on foundation agents for financial trading.
  • Two papers are accepted by ICLR 2024 on LLM-powered computer agents and finetuning LLM with online RL.
  • One paper is accepted by ICML 2023 on multi-agent reinforcement learning.
Research Highlights

Motivated by the goal of creating general-purpose agents that interact in both digital and physical worlds, my work primarily focused on using the Internet-scale knowledge within foundation models to generalize beyond available data and to unseen situations. Currently, I am interested in training vision-language-action models and generative interactive environments from actionless videos.

Publications ( show selected / show all by date / show all by topic )

Topics: Foundation Models for Decision Making
Past topics: Multi-Agent Reinforcement Learning / Others (*/†: indicates equal contribution.)

AgentStudio: A Toolkit for Building General Virtual Agents
Longtao Zheng*, Zhiyuan Huang*, Zhenghai Xue, Xinrun Wang, Bo An, Shuicheng Yan

NeurIPS 2024 Workshop on Open-World Agents Paper / Project Page / Code / Data

A trinity of environments, tools, and benchmarks for general virtual agents

Cradle: Empowering Foundation Agents Towards General Computer Control
Cradle Team

NeurIPS 2024 Workshop on Open-World Agents Paper / Project Page / Code

An agent that can play AAA video games

A Multimodal Foundation Agent for Financial Trading: Tool-Augmented, Diversified, and Generalist
Wentao Zhang, Lingxuan Zhao, Haochong Xia, Shuo Sun, Jiaze Sun, Molei Qin, Xinyi Li, Yuqing Zhao, Yilei Zhao, Xinyu Cai, Longtao Zheng, Xinrun Wang, Bo An

KDD 2024 Paper

The first multimodal agent for financial trading

Synapse: Trajectory-as-Exemplar Prompting with Memory for Computer Control
Longtao Zheng, Rundong Wang, Xinrun Wang, Bo An

ICLR 2024 Paper / Project Page / Code

A computer agent with state abstraction, trajectory prompting, and memory

True Knowledge Comes from Practice: Aligning Large Language Models with Embodied Environments via Reinforcement Learning
Weihao Tan, Wentao Zhang, Shanqi Liu, Longtao Zheng, Xinrun Wang, Bo An

ICLR 2024 Paper / Code

Train LLMs with online RL in embodied environments

Controlling Type Confounding in Ad Hoc Teamwork with Instance-wise Teammate Feedback Rectification
Dong Xing, Pengjie Gu, Qian Zheng, Xinrun Wang, Shanqi Liu, Longtao Zheng, Bo An, Gang Pan

ICML 2023 Paper

A causality-based solution to deal with type confounding in ad hoc teamwork

Multi-Agent Multi-Game Entity Transformer: Towards Generalist Models in MARL
Rundong Wang, Weixuan Wang, Xianhan Zeng, Liang Wang, Zhengjie Lian, Yiming Gao, Feiyu Liu, Siqin Li, Xianliang Wang, Qiang Fu, Wei Yang, Lanxiao Huang, Longtao Zheng, Zinovi Rabinovich, Bo An

DAI 2024 Paper

A generalist transformer for Honor of Kings, Starcraft II, and Neural MMO

Towards Skilled Population Curriculum for Multi-Agent Reinforcement Learning
Rundong Wang*, Longtao Zheng*, Wei Qiu, Bowei He, Bo An, Zinovi Rabinovich, Yujing Hu, Yingfeng Chen, Tangjie Lv, Changjie Fan

arXiv Preprint Paper / Code

Autocurricula for MARL in complex sparse-reward environments like Google Football

Contextual Data Cleaning with Ontology Functional Dependencies
Zheng Zheng, Longtao Zheng, Morteza Alipourlangouri, Fei Chiang, Lukasz Golab, Jaroslaw Szlichta, Sridevi Baskaran

ACM Journal of Data and Information Quality (JDIQ) Paper