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, 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 on multimodal digital agent research. I previously interned at SenseTime Research Singapore and Tencent Map, Beijing.

I am interested in the interaction and the productivity brought by digital (computer-use/GUI) agents. Currently I am focusing on game agents, as games offer ideal environments with rich visual observations, long-horizon goals, and latency-sensitive interaction. The improvement on perception, reasoning, and control will benefit the future embodiment of digital agents.

Email: yyyangwhu@gmail.com

Updates

[2026.09] GameWorld received the Best Paper Runner-Up Award at the MDA Workshop @ ECCV 2026 and invited to give a spotlight talk!
[2026.06] GameWorld was accepted to ECCV 2026🍻 and used as a multimodal interaction benchmark in Seed-2.1 Model Card.
[2026.04] We released GameWorld, a standardized benchmark for game agents. Check Project Page, Technical Report, and GitHub.
[2026.03] Our efficient UI grounding work FocusUI was accepted by CVPR 2026 with Compute Transparency Champion award.
[2025.09] Our slide generation work SlideTailor was accepted by AAAI 2026.
[2025.01] Started Ph.D. in School of Computing, NUS.
[2024.06] Our JPEG restoration work DCTransformer was published in IEEE TIP.
[2024.03] Our desktop GUI agent framework AssistGUI was accepted by CVPR 2024.
[2023.09] I joined ShowLab, NUS and started my internship at SenseTime Research Singapore.
[2023.08] Started my Master study in Data Science and Machine Learning at Faculty of Science, NUS.

Publications (*equal contribution)

	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, Best Paper Runner-Up Award, Multimodal Digital Agents Workshop @ ECCV 2026 📃 paper \| 🌐 project page \| 🏠 github \| ▶️ demo \| 🖼️ poster Standardizes the evaluation of game agents (both CUA and generalist agent interfaces) with 34 games and 170 tasks, using browser-based sandbox execution and state-verifiable scoring.
	FocusUI:Efficient UI Grounding via Position-Preserving Visual Token Selection Mingyu Ouyang, Kevin Qinghong Lin, Mike Zheng Shou, Hwee Tou Ng CVPR 2026 · CVPR Compute Transparency Champion award 📃 paper \| 🌐 project page \| 🏠 github \| 🤗 dataset \| 🖼️ poster Selects instruction-relevant UI tokens while preserving positional continuity, improving grounding accuracy and efficiency under aggressive token reduction.
	SlideTailor:Personalized Presentation Slide Generation for Scientific Papers Wenzheng Zeng, Mingyu Ouyang, Langyuan Cui*, Hwee Tou Ng AAAI 2026 📃 paper \| 🏠 github \| 🤗 dataset Generates preference-aligned editable 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 Techical Report 2024, GitHub 2k stars 📃 paper \| 🌐 project page \| 🏠 github An out-of-the-box framework for desktop GUI agents, supporting both Windows and macOS.
	AssistGUI:Task-Oriented PC Graphical User Interface Automation D. Gao, L. Ji, Z. Bai, M. Ouyang, P. Li, D. Mao, Q. Wu, W. Zhang, P. Wang, X. Guo, H. Wang, L. Zhou, Mike Z. Shou CVPR 2024 📃 paper \| 🌐 project page A Windows 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 IEEE Transactions on Image Processing 2024 📃 paper \| 🏠 github Recovers JPEG-quantized DCT coefficients with joint spatial-frequency modeling across varied compression quality factors.

Education

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

Experiences

Mar. 2025 - Oct. 2025	UseIt AI
Dec. 2023 - Oct. 2024	SenseTime Research Singapore
Mar. 2022 - Oct. 2022	Tencent Map, Beijing

Miscellaneous

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

	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, Best Paper Runner-Up Award, Multimodal Digital Agents Workshop @ ECCV 2026 📃 paper \| 🌐 project page \| 🏠 github \| ▶️ demo \| 🖼️ poster Standardizes the evaluation of game agents (both CUA and generalist agent interfaces) with 34 games and 170 tasks, using browser-based sandbox execution and state-verifiable scoring.
	FocusUI:Efficient UI Grounding via Position-Preserving Visual Token Selection Mingyu Ouyang, Kevin Qinghong Lin, Mike Zheng Shou, Hwee Tou Ng CVPR 2026 · CVPR Compute Transparency Champion award 📃 paper \| 🌐 project page \| 🏠 github \| 🤗 dataset \| 🖼️ poster Selects instruction-relevant UI tokens while preserving positional continuity, improving grounding accuracy and efficiency under aggressive token reduction.
	SlideTailor:Personalized Presentation Slide Generation for Scientific Papers Wenzheng Zeng, Mingyu Ouyang, Langyuan Cui*, Hwee Tou Ng AAAI 2026 📃 paper \| 🏠 github \| 🤗 dataset Generates preference-aligned editable 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 Techical Report 2024, GitHub 2k stars 📃 paper \| 🌐 project page \| 🏠 github An out-of-the-box framework for desktop GUI agents, supporting both Windows and macOS.
	AssistGUI:Task-Oriented PC Graphical User Interface Automation D. Gao, L. Ji, Z. Bai, M. Ouyang, P. Li, D. Mao, Q. Wu, W. Zhang, P. Wang, X. Guo, H. Wang, L. Zhou, Mike Z. Shou CVPR 2024 📃 paper \| 🌐 project page A Windows 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 IEEE Transactions on Image Processing 2024 📃 paper \| 🏠 github Recovers JPEG-quantized DCT coefficients with joint spatial-frequency modeling across varied compression quality factors.