Tao Yu (俞涛)

I am a Researcher at LimX Dynamics, which I joined in March 2025, specializing in reinforcement learning and humanoid control. My research interests span reinforcement learning, multimodal models, and AI-generated content (AIGC). My ultimate goal is to create intelligent agents capable of perceiving, reasoning, and acting like humans.

Prior to my current role, I served as a Senior Researcher at Tencent, focusing on reinforcement learning and game AI, after earning my Ph.D. in 2023 from the University of Science and Technology of China (USTC) under the supervision of Professors Zhibo Chen and Weiping Li. I also worked as a Research Student at the National University of Singapore (NUS) from 2022 to 2023, collaborating with Prof. Xinchao Wang. Earlier, I was a Research Intern at Microsoft Research Asia (MSRA) from 2020 to 2022, mentored by Dr. Cuiling Lan and Professor Wenjun Zeng.

I’m passionate about tackling exciting challenges through innovative projects. One of my recent works, Inpaint Anything, has garnered over 7k stars on GitHub.

If you’re interested in my research, collaboration opportunities, or internships in humanoid robotics, please don’t hesitate to reach out via email at yutao666@mail.ustc.edu.cn.

Selected Publications

Paper Image 1
Inpaint Anything: Segment Anything Meets Image Inpainting
Tao Yu, Runseng Feng, Ruoyu Feng, Jinming Liu, Xin Jin, Wenjun Zeng, Zhibo Chen
Technical Report 2023
Code
Paper Image 2
Task Residual for Tuning Vision-Language Models
Tao Yu*, Zhihe Lu*, Xin Jin, Zhibo Chen, Xinchao Wang
CVPR 2023
Code
Paper Image 2
Mask-based Latent Reconstruction for Reinforcement Learning
Tao Yu*, Zhizheng Zhang*, Cuiling Lan, Yan Lu, Zhibo Chen
NeurIPS 2022
Code
Paper Image 2
PlayVirtual: Augmenting Cycle-Consistent Virtual Trajectories for Reinforcement Learning
Tao Yu, Cuiling Lan, Wenjun Zeng, Mingxiao Feng, Zhizheng Zhang, Zhibo Chen
NeurIPS 2021
Code
Paper Image 2
Region Normalization for Image Inpainting
Tao Yu, Zongyu Guo, Xin Jin, Shilin Wu, Zhibo Chen, Weiping Li, Zhizheng Zhang, Sen Liu
AAAI 2020
Code
Paper Image 2
Local Patch Autoaugment with Multi-Agent Collaboration
Shiqi Lin*, Tao Yu*, Ruoyu Feng, Xin Li, Xiaoyuan Yu, Lei Xiao, Zhibo Chen
IEEE Transactions on Multimedia 2023
Code
Paper Image 2
GraphIQA: Learning Distortion Graph Representations for Blind Image Quality Assessment
Simeng Sun*, Tao Yu*, Jiahua Xu, Wei Zhou, Zhibo Chen
IEEE Transactions on Multimedia 2023
Code

(* indicates equal contribution)

Selected Awards and Competitions

  • 2024 Tencent Knowledge Award
  • 2023 Outstanding Ph.D Graduate of USTC
  • 2023 AI Rising Star Award
  • 2022 Champion of CVPR 2022 Challenge on Learned Image Compression (Perceptual Track)
  • 2019 Champion of ICME 2019 Grand Challenge on Learning-Based Image Inpainting
  • 2017 Honorable Prize of International Mathematical Contest in Modeling
  • 2016 First Prize of China Undergraduate Mathematical Contest in Modeling

Service

Reviewer: NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, AAAI, TPAMI, IJCV, TMM, etc.