avatar

Zihao Wang

Researcher
ByteDance Seed
zhwang1694 (at) gmail.com


About Me

I’m a researcher at ByteDance Seed. I received my Ph.D. degree from the Institute for Artificial Intelligence, Peking University (PKU) in 2026, advised by Prof. Yitao Liang and Prof. Yaodong Yang. I also collaborated closely with Dr. Xiaojian Ma and Prof. Anji Liu. My research centers on building large language models for generalist agents, computer-use agents, and real-time game agents.

News

Technical Reports

As a contributor of the ByteDance Seed team.

  1. Tech Report
    Seed2.1: Advancing AI Productivity
    ByteDance Seed
    Technical Report, 2026.
    Blog

  2. Tech Report
    Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity
    ByteDance Seed
    Technical Report, 2026.
    Project Paper

  3. Tech Report
    Seed1.8 Model Card: Towards Generalized Real-World Agency
    ByteDance Seed
    Technical Report, 2025.
    Project Paper

  4. Tech Report
    Seed1.5-VL Technical Report
    ByteDance Seed
    Technical Report, 2025.
    Project Paper API

Selected Publications

(* denotes equal contribution. For the full publication list, please see my Google Scholar.)

  1. ICML
    OpenHA: A Series of Open-Source Hierarchical Agentic Models in Minecraft
    Zihao Wang, Muyao Li, Kaichen He, Xiangyu Wang, Zhancun Mu, Anji Liu, Yitao Liang
    International Conference on Machine Learning (ICML), 2026.
    Paper Code Model

  2. arXiv
    Game-TARS: Pretrained Foundation Models for Scalable Generalist Multimodal Game Agents
    Bytedance Seed
    arXiv preprint, 2025.
    Project Paper

  3. arXiv
    UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning
    Bytedance Seed
    arXiv preprint, 2025.
    Project Paper

  4. ACL Findings
    JARVIS-VLA: Post-Training Large-Scale Vision Language Models to Play Visual Games with Keyboards and Mouse
    Muyao Li*, Zihao Wang*, Kaichen He, Xiaojian Ma, Yitao Liang
    Findings of the Association for Computational Linguistics (ACL Findings), 2025.
    Project Paper Code Dataset Model

  5. NeurIPS
    OmniJARVIS: Unified Vision-Language-Action Tokenization Enables Open-World Instruction Following Agents
    Zihao Wang, Shaofei Cai, Zhancun Mu, Haowei Lin, Ceyao Zhang, Xuejie Liu, Qing Li, Anji Liu, Xiaojian Ma, Yitao Liang
    Conference on Neural Information Processing Systems (NeurIPS), 2024.
    Project Paper Twitter

  6. NeurIPS Workshop
    RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Horizon Generation
    Zihao Wang, Anji Liu, Haowei Lin, Jiaqi Li, Xiaojian Ma, Yitao Liang
    Conference on Neural Information Processing Systems (NeurIPS) Workshop, 2024.
    Project Demo Paper Code Twitter

  7. T-PAMI
    JARVIS-1: Open-World Multi-task Agents with Memory-Augmented Multimodal Language Models
    Zihao Wang, Shaofei Cai, Anji Liu, Yonggang Jin, Jinbing Hou, Bowei Zhang, Haowei Lin, Zhaofeng He, Zilong Zheng, Yaodong Yang, Xiaojian Ma, Yitao Liang
    IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), 2024.
    Project Paper Code Twitter Media

  8. NeurIPS
    Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents
    Zihao Wang, Shaofei Cai, Guanzhou Chen, Anji Liu, Xiaojian Ma, Yitao Liang
    Conference on Neural Information Processing Systems (NeurIPS), 2023.
    Paper Code Twitter Best Paper Award, ICML 2023 TEACH Workshop

Experience

Honors & Awards

Talks

Services

Workshop Organizer

Conference Reviewers

Teaching Assistant


Powered by Jekyll and Minimal Light theme.