Liang-Chieh (Jay) Chen- Home Page

Liang-Chieh Chen
Research Scientist and Manager, TikTok, Los Angeles

Email: lcchen at cs dot ucla dot edu

About Me

I am currently a Research Scientist and Manager at TikTok with a focus on Computer Vision and Deep Learning.

I am widely recognized for my contributions to the DeepLab series (co-developed with George Papandreou): DeepLabv1, DeepLabv2, DeepLabv3, DeepLabv3+. Since December 2014, our introduced atrous convolution (also known as convolution with holes or dilated convolution) has become a foundational technique for dense prediction tasks.

Influential DeepLab derivatives include Auto-DeepLab, Panoptic-DeepLab (winning method of the Mapillary Vistas Panoptic Segmentation track at ICCV 2019, and top performer on Cityscapes leaderboards), Axial-DeepLab, ViP-DeepLab, MaX-DeepLab, and kMaX-DeepLab. Additionally, I am known for my collaborative work on MobileNetv2 and MobileNetv3, which have set new standards in efficient neural network design and search for mobile devices.

Prior to joining TikTok, I spent seven years as a Research Scientist at Google Research in Los Angeles (2016–2023). I earned my Ph.D. in Computer Science from the University of California, Los Angeles, in 2015, advised by Alan L. Yuille. I received my M.S. in Electrical and Computer Engineering from the University of Michigan, Ann Arbor.

Activities

Area Chair for ICCV 2019, 2025, CVPR 2020, 2023, 2024, 2025, ECCV 2020, 2024, NeurIPS 2022, 2024, ICML 2025.

Action Editor for TMLR.

Reviewer for CVPR, ECCV, ICCV, NeurIPS.

Selected Recent Publications

Democratizing Text-to-Image Masked Generative Models with Compact Text-Aware One-Dimensional Tokens
Dongwon Kim*, Ju He*, Qihang Yu*, Chenglin Yang, Xiaohui Shen, Suha Kwak, Liang-Chieh Chen
(*equal contribution)
Technical report
[preprint (arxiv: 2501.07730)] [project website] [code]

1.58-bit FLUX
Chenglin Yang, Celong Liu, Xueqing Deng, Dongwon Kim, Xing Mei, Xiaohui Shen, Liang-Chieh Chen
Technical report
[preprint (arxiv: 2412.18653)] [project website]

FlowAR: Scale-wise Autoregressive Image Generation Meets Flow Matching
Sucheng Ren, Qihang Yu, Ju He, Xiaohui Shen, Alan Yuille, Liang-Chieh Chen
Technical report
[preprint (arxiv: 2412.15205)] [code]

ViCaS: A Dataset for Combining Holistic and Pixel-level Video Understanding using Captions with Grounded Segmentation
Ali Athar, Xueqing Deng, Liang-Chieh Chen
Technical report
[preprint (arxiv: 2412.09754)] [project website] [code and dataset] [dataset (Hugging Face)]

Randomized Autoregressive Visual Generation
Qihang Yu, Ju He, Xueqing Deng, Xiaohui Shen, Liang-Chieh Chen
Technical report
[preprint (arxiv: 2411.00776)] [project website] [code]

MaskBit: Embedding-free Image Generation via Bit Tokens
Mark Weber, Lijun Yu, Qihang Yu, Xueqing Deng, Xiaohui Shen, Daniel Cremers, Liang-Chieh Chen
Transactions on Machine Learning Research (TMLR), December 2024. (Featured and Reproducibility Certifications)
[preprint (arxiv: 2409.16211)] [project website] [code]

Alleviating Distortion in Image Generation via Multi-Resolution Diffusion Models
Qihao Liu*, Zhanpeng Zeng*, Ju He*, Qihang Yu, Xiaohui Shen, Liang-Chieh Chen
(*equal contribution)
In Neural Information Processing Systems (NeurIPS), Vancouver, Canada, December 2024.
[preprint (arxiv: 2406.09416)] [project website] [code]

An Image is Worth 32 Tokens for Reconstruction and Generation
Qihang Yu*, Mark Weber*, Xueqing Deng, Xiaohui Shen, Daniel Cremers, Liang-Chieh Chen
(*equal contribution)
In Neural Information Processing Systems (NeurIPS), Vancouver, Canada, December 2024.
[preprint (arxiv: 2406.07550)] [project website] [code]