Haoxin Li     李昊昕

AI Engineer

Tencent, China

Email: lihaoxin05@gmail.com

About Me (CV)

I am currently an AI Engineer at Tencent. I obtained my Ph.D. degree from Nanyang Technological University, under the supervision of Prof. Boyang Li and Prof. Hanwang Zhang. I received my Master degree from Sun Yat-Sen University, under the supervision of Prof. Wei-Shi Zheng.

Education

• Nanyang Technological University: Aug., 2021 – Aug., 2025
  • Ph.D. in Computer Science and Engineering
• Sun Yat-Sen University: Aug., 2018 – June, 2021
  • M.Sc. in Information and Communication Engineering
• Sun Yat-Sen University: Aug., 2014 – June, 2018
  • B.E. in Electronic Engineering

Publications

• Haoxin Li, Yingchen Yu, Qilong Wu, Hanwang Zhang, Song Bai, Boyang Li. Learning to Animate Images from A Few Videos to Portray Delicate Human Actions. IEEE Winter Conference on Applications of Computer Vision (WACV), 2026. [arxiv][webpage][code][main paper][supplementary material]

• Haoxin Li, Boyang Li. Enhancing Vision-Language Compositional Understanding with Multimodal Synthetic Data. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025. [arxiv][code][main paper][supplementary material]

• Haoxin Li, Yuan Liu, Hanwang Zhang, Boyang Li. Mitigating and Evaluating Static Bias of Action Representations in the Background and the Foreground. IEEE International Conference on Computer Vision (ICCV), 2023. (Oral) [arxiv][code][main paper][supplementary material]

• Haoxin Li, Wei-Shi Zheng, Jianguo Zhang, Haifeng Hu, Jiwen Lu, Jian-Huang Lai. Egocentric Action Recognition by Automatic Relation Modeling. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2022. [paper][code]

• Haoxin Li, Wei-Shi Zheng, Yu Tao, Haifeng Hu, Jian-Huang Lai. Adaptive Interaction Modeling via Graph Operations Search. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. [arxiv][code][main paper][supplementary material]

• Haoxin Li, Yijun Cai, Wei-Shi Zheng. Deep Dual Relation Modeling for Egocentric Interaction Recognition. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019. [arxiv][code][main paper] [supplementary material]

• Shuowei Li, Haoxin Li, Wenda Chu, Yi Fang. How Large Language Models Balance Internal Knowledge with User and Document Assertions. Findings of the Association for Computational Linguistics: ACL 2026. [paper]

• Miaoyu Li, Haoxin Li, Zilin Du, Boyang Li. Diversify, Rationalize, and Combine: Ensembling Multiple QA Strategies for Zero-shot Knowledge-based VQA. Findings of the Association for Computational Linguistics: EMNLP 2024. [paper]

• Yi Zhong, Jia-Hui Pan, Haoxin Li, Wei-Shi Zheng. Weakly supervised action anticipation without object annotations. Frontiers of Computer Science, 2023. [paper]

• Jiaming Zhou, Kun-Yu Lin, Haoxin Li, Wei-Shi Zheng. Graph-Based High-Order Relation Modeling for Long-Term Action Recognition. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021. [paper]

• Yijun Cai, Haoxin Li, Jian-Fang Hu, Wei-Shi Zheng. Action Knowledge Transfer for Action Prediction with Partial Videos. 33rd AAAI Conference on Artificial Intelligence (AAAI), 2019. [paper]

• Shuosen Guan, Haoxin Li, Wei-Shi Zheng. Unsupervised Learning for Optical Flow Estimation Using Pyramid Convolution LSTM. IEEE International Conference on Multimedia and Expo (ICME), 2019. [paper]

Honors and Awards

• ActivityNet Large-Scale Activity Recognition Challenge 2018: Trimmed Event Recognition (Moments in Time Recognition Challenge), Rank: 1/12 in Mini Track, 10/29 in Full Track. [report]
• National Scholarship (1/264), by Minister of Education of China, 2015

Service

• Reviewer of CVPR2024, CVPR2025, CVPR2026
• Reviewer of ICCV2023, ICCV2025
• Reviewer of ECCV2024, ECCV2026
• Reviewer of NeurIPS2024, NeurIPS2025
• Reviewer of ICLR2025, ICLR2026
• Reviewer of AAAI2023, AAAI2024, AAAI2025, AAAI2026
• Reviewer of ACMMM2025, ACMMM2026