Kaiqi Liu

liukq04@163.com

I am a first-year Ph.D. student at Peking University, advised by Prof. Boxin Shi in the Camera Intelligence Lab (CI Lab). Prior to this, I received my B.E. in Software Engineering from Nanjing University with honors, with GPA ranked 1/216 (top 0.5%).

Research Interests

Generative AI (AIGC): Controllable Video Generation, Multi-Shot Video Generation, Audio-Visual Generation, and Multimodal Evaluation.
World Models (WM): Interactive World Models, Autoregressive Long Video Generation, Memory-Augmented Modeling.
Unified Understanding and Generation Models (UMM): Unified Image and Video Understanding and Generation, New Unified Video Generation Architectures.

Feel free to reach out! I’m always open to research collaborations and discussions! ✨

News

Jun 2026	One paper was accepted to ACM MM 2026.
May 2026	Two papers were accepted to ECCV 2026, with MAVIN selected as an Oral presentation.
Jan 2026	One paper was accepted to CVPR 2026.
May 2025	One paper was accepted to ICCV 2025.
Apr 2025	One paper was accepted to ICML 2025.

Selected Publications

ACM MM 2026

MILD: Multi-Layer Diffusion Strategy for Complex and Precise Multi-IP Aware Human Erasing

Jinghan Yu, Junhao Xiao, Zhiyuan Ma, Yue Ma, Kaiqi Liu, Yuhan Wang, Daizong Liu, Xianghao Meng, and Jianjun Li

Proceedings of the ACM International Conference on Multimedia (ACM MM), 2026

Website / Paper
ECCV 2026

MAVIN: Multi-Shot Audio-Visual Generation with Customized Narrative Control

Kaiqi Liu, Yunyao Mao, Ziqi Cai, Zheng Geng, Jing Wang, Qiulin Wang, Xintao Wang, Pengfei Wan, Kun Gai, Shuchen Weng, and Boxin Shi

Proceedings of the European Conference on Computer Vision (ECCV), 2026

ECCV 2026 Oral Presentation

Website / Paper
ECCV 2026

Video Generation Models Are Inherent Lighting Estimators

Ziqi Cai, Shuchen Weng, Kaiqi Liu, Zifeng Wang, Zhiquan Zhang, Minggui Teng, Han Jiang, and Boxin Shi

Proceedings of the European Conference on Computer Vision (ECCV), 2026

Website / Paper
CVPR 2026

STAGE: Storyboard-Anchored Generation for Cinematic Multi-shot Narrative

Peixuan Zhang, Zijian Jia, Kaiqi Liu, Shuchen Weng, Si Li, and Boxin Shi

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026

Paper / Code
ICCV 2025

Adapting In-Domain Few-Shot Segmentation to New Domains without Source Domain Retraining

Qi Fan, Kaiqi Liu, Nian Liu, Hisham Cholakkal, Rao Muhammad Anwer, Wenbin Li, and Yang Gao

Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2025

Paper / Code
ICML 2025

Improving LLMs for Recommendation with Out-Of-Vocabulary Tokens

Ting-Ji Huang, Jia-Qi Yang, Chunxu Shen, Kaiqi Liu, De-Chuan Zhan, and Han-Jia Ye

Proceedings of the International Conference on Machine Learning (ICML), 2025

Paper / Code