Siyuan Hu

Incoming CS Ph.D. Student, Nanyang Technological University

I am a Ph.D. student in Computer Science at MMLab, Nanyang Technological University, advised by Prof. Ziwei Liu. I obtained my Bachelor's degree in Computer Science from the same university.

Previously, I worked with Prof. Mike Shou at Show Lab, NUS, and Dr. Hongyuan Zhu at A*STAR.

My research focuses on agents and their environments, from digital to physical worlds.

Email: siyuan.hu.sg[AT]gmail.com

Updates

[2026/07] PaperDoctor received the Best Poster Award at AI for Science @ ICML 2026.
[2026/06] GameWorld was accepted to ECCV 2026.
[2026/02] ShowUI-π was accepted to CVPR 2026.
[2025/11] We released CUA-as-Judges, using computer-use agents as judges for generative user interfaces (LAW @ NeurIPS 2025).
[2024/10] AssistGPT received the Best Demo Paper Award at ACM MM 2024 HCMA Workshop.

Selected Publications (*equal contribution)

CVPR '26	ShowUI-π:Flow-based Generative Models as GUI Dexterous Hands Siyuan Hu, Kevin Qinghong Lin, Mike Zheng Shou CVPR 2026 📃 paper \| 🌐 project page Flow-based generative models for precise and dexterous GUI action execution.
ECCV '26	GameWorld:Towards Standardized and Verifiable Evaluation of Multimodal Game Agents Mingyu Ouyang, Siyuan Hu, Kevin Qinghong Lin, Hwee Tou Ng, Mike Zheng Shou ECCV 2026 📃 paper \| 🌐 project page \| 🏠 github Standardizes game agent evaluation across 34 browser games and 170 tasks with browser-based sandbox execution and state-verifiable scoring.
arXiv '25	CUA-as-Judges:Computer-Use Agents as Judges for Generative User Interface Kevin Qinghong Lin, Siyuan Hu, Linjie Li, Zhengyuan Yang, Lijuan Wang, Philip Torr, Mike Zheng Shou Preprint 2025, LAW @ NeurIPS 2025 📃 paper \| 🌐 project page \| 🏠 github \| 🤗 demo Employs computer-use agents as judges to evaluate generative user interfaces through real interaction.
ICMLW '26	PaperDoctor:Evidence-Grounded and Actionable Feedback for Scientific Papers in Progress Kevin Qinghong Lin, Siyuan Hu, Pan Lu, Yu Chen, Yanzhe Chen, Owen Queen, Yupeng Chen, Jialin Yu, Junchi Yu, Zifeng Ding, Yuanfeng Ji, Sheng Liu, Jindong Gu, Linjie Li, Mike Zheng Shou, Philip Torr, James Zou AI for Science @ ICML 2026 Best Poster Award Provides evidence-grounded and actionable feedback for scientific papers in progress.
MM '24	AssistGPT:Towards Multi-modal Agent for Human-Centric AI Assistant Difei Gao, Siyuan Hu, Zechen Bai, Kevin Qinghong Lin, Mike Zheng Shou ACM MM 2024, HCMA Workshop Best Demo Paper Award 📃 paper A multi-modal agent framework that plans and chains AI tools for human-centric assistance.

Open Source

Computer Use Out-of-the-box — creator and lead maintainer of one of the earliest computer-use agent frameworks. GitHub 2k stars
Awesome-GUI-Agent — main maintainer of the GUI agent paper list. GitHub 1.2k stars

Honors

2026	Best Poster Award AI for Science Workshop @ ICML
2026 - 2030	Nanyang President's Graduate Scholarship (NPGS)
2025 - 2026	Tinker Research Grant Thinking Machines Lab
2025	Show Lab Annual Award
2024	Best Demo Paper ACM MM Human-Centric Multimedia Analysis Workshop
2021 - 2023	NTU President Research Scholar
2020 - 2024	NTU Science and Technology Undergraduate Full Scholarship Dean's List (top 5%)

Education

2026 -	Nanyang Technological University Ph.D. in Computer Science
2020 - 2024	Nanyang Technological University B.Eng. in Computer Science

Miscellaneous

I am a beatmaker; I use a Roland SP-404MKII for beatmaking and live jamming, in jazz and lo-fi vibes.