I am Wei Li, a principal research scientist at Huawei Foundation Model Department. Currently, I am leading the Multimodal Generation Post-Training team. My research interests broadly encompass efficient systems for large language models and multimodal generation, and reinforcement learning for multimodal generation.

I received my M.S. degree from Tsinghua University in 2019, co-advised by Prof. Guoqiang Bai and IEEE Fellow Prof. Zhihua Wang. Since joining Huawei in 2019, I have collaborated closely with Dr.Xinghao Chen and Dr.Yunhe Wang

Driven by a passion for bridging cutting-edge algorithms with strict hardware constraints, I have pioneered multiple core technologies that empower intelligent imaging and generative models across tens of millions of flagship devices globally. I have published over 20 papers in top-tier AI venues (including multiple ICLR Spotlight, CVPR, ICML, ACL, and IEEE TIP articles) and hold more than 30 patents.

🤝 Join Our Team / Hiring Interns

We are actively seeking self-motivated interns to work on next-generation multimodal generation systems. If you are passionate about defining the future of ubiquitous generative AI and publishing at top-tier venues, please feel free to drop me an email with your CV.

🚀 News

===========
[2026.05] Three papers accepted by ICML 2026 (ES-COT, CSD, Extra-CoT).
[2026.04] One paper accepted by ACL 2026 (Dac-KL).
[2025.11] One paper accepted by IEEE TIP (Hi-Mamba).
[2025.05] One paper accepted by IEEE TIP. (LIPT).
[2025.01] One paper accepted by CVPR 2025.
[2025.01] Three papers accepted by ICLR 2025, including 2 Spotlight(CBQ, MipKD)！
[2024.11] Two papers accepted by AAAI 2025 (GIM, DCKD).
[2024.07] One paper accepted by ECCV 2024 (PQ-SAM).
[2024.02] One paper accepted by CVPR 2024 (SAM4IR).
[2023.09] One paper accepted by NeurIPS 2024 (Genimage).
[2023.02] One paper accepted by CVPR 2023 (RefSR-NeRF).
[2021.11] One paper accepted by AAAI 2022 (OoDHDR-codec).

Selected Publications

Knowledge distillation with multi-granularity mixture of priors for image super-resolution

Simiao Li, Yun Zhang, Wei Li^†, Hanting Chen, Wenjia Wang, Bingyi Jing, Shaohui Lin, Jie Hu

^†Project Leader & Co-first author, ICLR 2025 (Spotlight)   [Paper]

LIPT: Latency-Aware Image Processing Transformer

Junbo Qiao, Wei Li^†, Haizhen Xie, Hanting Chen, Jie Hu, Shaohui Lin, Jungong Han

^†Project Leader & Co-first author, IEEE TIP 2025   [Paper]

AugKD: Ingenious Augmentations Empower Knowledge Distillation for Image Super-Resolution

Yun Zhang, Wei Li^†, Simiao Li, Hanting Chen, Zhijun Tu, Wenjia Wang, Bingyi Jing, Shaohui Lin, Jie Hu

^†Project Leader & Co-first author, ICLR 2025   [Paper]

RefSR-NeRF: Towards High Fidelity and Super Resolution View Synthesis

Xudong Huang, Wei Li^†, Jie Hu, Hanting Chen, Yunhe Wang

^†Project Leader & Co-first author, CVPR 2023   [Paper]

RDDM: PRACTICING RAW DOMAIN DIFFUSION MODEL FOR REAL-WORLD IMAGE RESTORATION

Yan Chen, Yi Wen, Wei Li^†, Junchao Liu, Yong Guo, Jie Hu, Xinghao Chen

^†Project Leader, Arxiv 2025   [Paper]

Multi-Granularity Semantic Revision for Large Language Model Distillation

Xiaoyu Liu1, Yun Zhang, Wei Li^†, Xudong Huang, Hanting Chen, Yehui Tang, Jie Hu, Zhiwei Xiong, and Yunhe Wang

^†Project Leader, ACL 2026   [Paper]

CBQ: CROSS-BLOCK QUANTIZATION FOR LARGE LANGUAGE MODELS

Xin Ding, Xiaoyu Liu, Zhijun Tu, Yun Zhang, Wei Li^†, Jie Hu, Hanting Chen, Yehui Tang, Zhiwei Xiong, BaoqunYin, YunheWang

^†ICLR 2025 (Spotlight)   [Paper]