Jiahang Zhang

ABOUT ME

Hi, I'm Zhang Jiahang, currently pursuing the Ph.D. degree at STRUCT group, WICT, guided by Professor Jiaying Liu. My research interest lies in human-centric video motion understanding, especially via multimodal language models.

Experience

09/2019 - 07/2023: Bachelor major in Computer Science and Technology, EECS, Peking University.

09/2024 - 08/2025: Visiting student at Peng Cheng Laboratory, Shen Zhen. Research on Feature Coding and Compression for Vision Foundation Models.

12/2025 - Present: Research intern at Kling Team (Kuaishou Technology) on Multimodal Large Language Models, especially for Video Understanding.

Publications

* indicates equal contribution.

- Jiahang Zhang, Yushuo Guan, Yuanxing Zhang, Pengfei Wan, and Jiaying Liu. "Iris: Empowering Video MLLMs with High-Frequency Pose Priors via Spatiotemporal Binding", NeurIPS, Sydney, Australia, December 2026.

- Jiahang Zhang, Wenhan Yang, Minghao Liu, and Jiaying Liu. "Towards Reconfigurable Visual Feature Compression", ECCV, Malmö, Sweden, September 2026.

- Bozhou Li*, Jiahang Zhang*, et. al. "Human-Centric Image Captioning with Subject-Centered Spatial Understanding", ACM MM, Rio de Janeiro, Brazil, November 2026.

- Jiahang Zhang, Lilang Lin, Shuai Yang, Jiaying Liu. "Self-Supervised Skeleton-Based Action Representation Learning: A Benchmark and Beyond", IJCV, 2025.

- Lilang Lin, Jiahang Zhang, Jiaying Liu. "Self-Supervised Skeleton Representation Learning via Actionlet Contrast and Reconstruct", IEEE TPAMI, 2025.

- Jiahang Zhang, Lilang Lin, Shuai Yang, Jiaying Liu. "SGAR: Structural Generative Augmentation for 3D Human Motion Retrieval", NeurIPS, SAN DIEGO, USA, December 2025.

- Jiahang Zhang, Lilang Lin, Jiaying Liu. "Shap-Mix: Shapley Value Guided Mixing for Long-Tailed Skeleton Based Action Recognition", IJCAI, Jeju, Korea, August 2024.

- Jiahang Zhang*, Lilang Lin*, Zejia Fan, Wenjing Wang, Jiaying Liu. "S5Mars: Semi-Supervised Learning for Mars Semantic Segmentation", IEEE TGRS, 2024.

- Lilang Lin*, Jiahang Zhang*, Jiaying Liu. "Mutual Information Driven Equivariant Contrastive Learning for 3D Action Representation Learning", IEEE TIP, 2024.

- Jiahang Zhang, Lilang Lin, Jiaying Liu. "Prompted Contrast with Masked Motion Modeling: Towards Versatile 3D Action Representation Learning", ACM MM, Ottawa, Canada, October 2023.

- Lilang Lin, Jiahang Zhang, Jiaying Liu. "Actionlet-Dependent Contrastive Learning for Unsupervised Skeleton-Based Action Recognition", CVPR, Highlight, Vancouver, Canada, June, 2023.

- Jiahang Zhang, Lilang Lin, Jiaying Liu. "Hierarchical Consistent Contrastive Learning for Skeleton-Based Action Recognition with Growing Augmentations", AAAI, Oral, WASHINGTON, DC, USA, February 2023.

Others

• Ranked in the top 7% of GPA during undergraduate study.

• Merit Student Award, Academic Excellence Award, Arawana Scholarship, The Second Prize of PKU Scholarship, Outstanding Graduate in Peking University.

• First prize of Wang Xuan Scholarship in WICT 2023.

• National Scholarship in PKU 2024.