I am an M.S. student in Artificial Intelligence at Tsinghua University. Previously, I received my B.Eng. in Artificial Intelligence from Dalian University of Technology.

My research focuses on building Universal Digital Agents (UDAs), with an emphasis on Computer-Use Agents. I study how agents can understand and operate across digital environments, including graphical interfaces, terminals, and code. I am particularly interested in long-horizon reasoning, post-training, and Agentic RL for UDAs.

I aim to make computer use a core capability of frontier models and to bring UDAs into everyday life. Computers serve as the core medium for these agents. By operating digital systems and connecting them with real-world tasks, they can improve accessibility, extend human capabilities, and create meaningful impact in the physical world. I believe UDAs represent an important path toward AGI.

I am currently a research intern with the Qwen Team at Alibaba, where I am advised by Que Shen. Previously, I interned at Huawei Multimodal Lab and collaborated remotely with Microsoft Research Asia.

If you would like to talk about research or life, feel free to reach out. I am open to research collaborations and opportunities.

News

  • 2026.08 WeaveBench and ST-Lite accepted by EMNLP 2026.
  • 2026.07 Started a research internship with the Qwen Team at Alibaba.
  • 2026.06 Released WeaveBench, a benchmark for hybrid-interface Computer-Use Agents.
  • 2026.04 Started a research internship at Huawei Multimodal Lab.
  • 2026.02 Released ST-Lite for efficient long-horizon GUI agents.
  • 2026.01 Began a remote research collaboration with Microsoft Research Asia.

Research

Toward universal agents that can perceive, reason, and act across diverse digital interfaces.

Universal Digital Agents

Research spanning GUI interaction, terminal and tool use, code execution, long-horizon reasoning, evaluation, and efficient deployment.

WeaveBench task examples

WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces

Bowen Zhou, et al.

First Author EMNLP 2026

A benchmark of 114 real-world tasks across eight domains that require agents to interleave GUI observation with CLI and code execution; the best evaluated system reaches only 41.2% success.

ST-Lite overview

ST-Lite: Training-Free KV Cache Compression with Spatio-Trajectory Guidance for Long-Horizon GUI Agents

Bowen Zhou, et al.

First Author EMNLP 2026

A training-free KV cache compression method tailored to GUI interaction traces, matching or exceeding Full Cache accuracy at a 20% budget while delivering up to 2.35× decoding speedup.

GUIPruner method overview

GUIPruner: Spatio-Temporal Token Pruning for Efficient High-Resolution GUI Agents

Bowen Zhou, et al.

First Author arXiv 2026

A training-free framework that combines temporal-adaptive resolution with structure-aware visual-token pruning, achieving a 3.4× FLOPs reduction and a 3.3× vision-encoding speedup while retaining over 94% of the original performance.

Experience

Qwen Team, Alibaba Group Jul 2026 – Present

Research Intern · Base Model Team · Advised by Que Shen

Research on foundation models, multimodal reasoning, and generalist agents for complex digital environments.

Huawei Multimodal Lab Apr 2026 – Jul 2026

Research Intern · Base Model Team

Multimodal foundation-model research, scalable training, and research prototyping.

Microsoft Research Asia Jan 2026 – May 2026

Research Collaboration · Remote

Hybrid-interface Computer-Use Agents and the WeaveBench benchmark.

Skills

Research Universal Digital Agents · Agentic RL · Multimodal Foundation Models · Inference Optimization
Languages Python · C/C++ · LaTeX
Frameworks PyTorch · veRL · slime · vLLM · SGLang · Transformers
Systems Docker · Kubernetes · VM-based agent environments · large-scale evaluation

Education

  • Tsinghua University — M.S. in Artificial Intelligence Present
  • Dalian University of Technology — B.Eng. in Artificial Intelligence