Mingyu Ouyang

CS Ph.D. Student, National University of Singapore

I am a second-year Ph.D. student in Computer Science at the National University of Singapore, co-advised by Prof. Hwee Tou Ng and Prof. Mike Zheng Shou. I completed my Master's at NUS, and obtained my Bachelor's degree from Wuhan University.

I am with Show Lab, NUS and ARC Lab, Tencent IEG on multimodal digital agent research. I previously interned at SenseTime Research Singapore and Tencent Map, Beijing.

I am interested in the interaction and productivity gains enabled by digital (computer-use/game) agents. Recently, I am focusing on game environments, as games offer rich visual observations, long-horizon goals, and dynamic mechanics. Improvements in perception, reasoning, and control in games will support the future embodiment of digital agents.

Email: yyyangwhu@gmail.com

Updates

10/2026 We released Player-1, a post-training framework for learning game control from synthetic experience. Check the paper, project page, and GitHub.
09/2026 GameWorld received the Best Paper Runner-Up Award and invited to give a spotlight talk at the MDA Workshop @ ECCV 2026.
08/2026 Received the Research Achievement Award from NUS School of Computing.
06/2026 GameWorld was accepted to ECCV 2026 and used in Seed-2.1 Model Card.
04/2026 We released GameWorld, a standardized benchmark for game agents. Check Project Page, Technical Report, and GitHub.
03/2026 FocusUI was accepted to CVPR 2026 and received the Compute Transparency Champion Award.

Earlier updates (2023–2025)

09/2025 SlideTailor was accepted by AAAI 2026.
01/2025 Started Ph.D. in School of Computing, NUS.
06/2024 DCTransformer was accepted by IEEE Transactions on Image Processing.
03/2024 AssistGUI was accepted by CVPR 2024.
09/2023 Joined Show Lab, NUS and started an internship at SenseTime Research Singapore.
08/2023 Started Master's in Data Science and Machine Learning at Faculty of Science, NUS.

Publications (* equal contribution; † corresponding author)

	Player-1: Learning to Play from Synthetic Experience Mingyu Ouyang, Yiran Wang, Siyuan Hu, Xingyilang Yin, Huiqiang Sun, Guangzhi Wang, Junfu Pu, Cheng Cheng, Honghao Chen, Ying Shan, Mike Zheng Shou^†, Hwee Tou Ng^†M. Ouyang, Y. Wang, S. Hu, X. Yin, H. Sun, G. Wang, J. Pu, C. Cheng, H. Chen, Y. Shan, M. Z. Shou^†, H. T. Ng^† Under review, 2026 📃 paper 🌐 project page 🏠 github ▶️ introduction Synthetic environments improve transferable agent capabilities: An agentic post-training framework that learns general game control beyond supervised demonstrations.
	GameWorld:Towards Standardized and Verifiable Evaluation of Multimodal Game Agents Mingyu Ouyang, Siyuan Hu, Kevin Qinghong Lin, Hwee Tou Ng^†, Mike Zheng Shou^†M. Ouyang, S. Hu, K. Q. Lin, H. T. Ng^†, M. Z. Shou^† ECCV 2026, Best Paper Runner-Up Award, Multimodal Digital Agents Workshop @ ECCV 2026 📃 paper 🌐 project page 🏠 github ▶️ demo 🖼️ poster 📊 slides Can VLMs play games as humans do? GameWorld evaluates game agents (both computer-use agents and general-purpose agents) with 34 games and 170 tasks.
	FocusUI:Efficient UI Grounding via Position-Preserving Visual Token Selection Mingyu Ouyang, Kevin Qinghong Lin, Mike Zheng Shou^†, Hwee Tou Ng^†M. Ouyang, K. Q. Lin, M. Z. Shou^†, H. T. Ng^† CVPR 2026, CVPR Compute Transparency Champion Award 📃 paper 🌐 project page 🏠 github 🤗 dataset 🖼️ poster Makes UI agents' perception more efficient: FocusUI selects instruction-relevant visual tokens for low-cost UI grounding.
	SlideTailor:Personalized Presentation Slide Generation for Scientific Papers Wenzheng Zeng, Mingyu Ouyang, Langyuan Cui, Hwee Tou Ng^†W. Zeng, M. Ouyang, L. Cui, H. T. Ng^† AAAI 2026 📃 paper 🏠 github 🤗 dataset Generates preference-aligned slides for scientific papers.
	The Dawn of GUI Agent:A Preliminary Case Study with Claude 3.5 Computer Use Siyuan Hu, Mingyu Ouyang, Difei Gao, Mike Zheng Shou^†S. Hu, M. Ouyang, D. Gao, M. Z. Shou^† Technical Report 2024, GitHub 2k stars 📃 paper 🌐 project page 🏠 github An out-of-the-box framework for desktop GUI agents, supporting multiple OSs and models.
	AssistGUI: Task-Oriented PC Graphical User Interface Automation Difei Gao, Lei Ji, Zechen Bai, Mingyu Ouyang, Peiran Li, Dongxing Mao, Qinchen Wu, Weichen Zhang, Peiyi Wang, Xiangwu Guo, Hengxu Wang, Luowei Zhou, Mike Zheng Shou^†D. Gao, L. Ji, Z. Bai, M. Ouyang, P. Li, D. Mao, Q. Wu, W. Zhang, P. Wang, X. Guo, H. Wang, L. Zhou, M. Z. Shou^† CVPR 2024 📃 paper 🌐 project page A desktop GUI agent framework for task decomposition, GUI parsing, action generation, and reflection.
	JPEG Quantized Coefficient Recovery via DCT Domain Spatial-Frequential Transformer Mingyu Ouyang, Zhenzhong Chen^†M. Ouyang, Z. Chen^† IEEE Transactions on Image Processing 2024 📃 paper 🏠 github Recovers JEPG-quantized DCT coefficients with spatial-frequency modeling.

Education

2025 - Present	National University of Singapore Ph.D. in Computer Science
2023 - 2024	National University of Singapore M.Sc. in Data Science and Machine Learning
2019 - 2023	Wuhan University B.Sc. in Geospatial Informatics and Digitalized Technology

Experiences

2026	ARC Lab, Tencent IEG Research Intern (Project Up)
2025	UseIt AI Agent Dev
2024	SenseTime Research Singapore Research Intern
2022	Tencent Map Algorithm Dev Intern

Miscellaneous

I enjoy football games (Visca Barça! 💙❤️). I played for and served as the captain of the Wuhan University football team, where we won the championship at the 16th Hubei Province Games in 2022. "Salid y disfrutad."