Youwei Liang

I am a PhD student in the Department of Electrical and Computer Engineering at the University of California, San Diego. My PhD advisor is Prof. Pengtao Xie. Prior to that, I was an undergraduate student studying Information and Computing Science at SCAU, where I had the fortune to work with Prof. Dong Huang. I also spent time as a research intern at Tencent AI Lab, Meta AI, and Google Research.

News

• [04/2024] One paper was accepted as an Oral to CVPR 2024.
• [04/2023] One paper was accepted to ICML 2023.
• [01/2022] One paper was accepted as a spotlight at ICLR 2022.
• [12/2021] I co-organized an NeurIPS 2021 Workshop: Self-Supervised Learning - Theory and Practice.

Research Interests

I am broadly interested in machine learning, especially unsupervised/self-supervised learning such as generative models (e.g., GAN and VAE), contrastive/predictive learning (e.g., MoCo, BYOL, and CARE), and optimization.

I am currently working on accelerating the training and inference of deep learning models, with an emphasis on Vision Transformers. My previous projects span various aspects in machine learning, from multi-view clustering to theories of the Lipschitz constants and adversarial robustness of CNN, to self-supervised learning. If you are interested in my works, please feel free to contact me.

Selected Publications

Rich Human Feedback for Text-to-Image Generation
Youwei Liang, Junfeng He, Gang Li, Peizhao Li, Arseniy Klimovskiy, Nicholas Carolan, Jiao Sun, Jordi Pont-Tuset, Sarah Young, Feng Yang, Junjie Ke, Krishnamurthy Dj Dvijotham, Katherine M. Collins, Yiwen Luo, Yang Li, Kai J Kohlhoff, Deepak Ramachandran, and Vidhya Navalpakkam
The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2024 Oral). [arXiv]

Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations
Youwei Liang, Chongjian Ge, Zhan Tong, Yibing Song, Jue Wang, and Pengtao Xie
International Conference on Learning Representations (ICLR 2022 Spotlight). [OpenReview] [Code]

Consistency Meets Inconsistency: A Unified Graph Learning Framework for Multi-view Clustering
Youwei Liang, Dong Huang, and Chang-Dong Wang
IEEE International Conference on Data Mining (ICDM 2019). [PDF] [Code] [Slides] [YouTube]

More Publications

Professional Services

Reviewer

• ICML 2022, NeurIPS 2022, ICASSP 2023, ICLR 2023, CVPR 2023, ICML 2023, ICCV 2023, NeurIPS 2023
• IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)