Portrait of Jiajun Chen

Hello, Iā€™m

Jiajun Chen

Research Intern @ HKU MMLab

I am a research intern under the guidance of Prof. Xihui Liu at HKU MMLab, where I focus on embodied AI, egocentric video generation, and game world models. I'm currently open to industry internship opportunities and research collaborations. If you are interested in my work and would like to discuss potential collaborations, feel free to contact me via email! šŸ„°

I completed my bachelor's degree in Data Science and Big Data Technology at Beijing University of Posts and Telecommunications. During my senior year, I visited the National University of Singapore and worked as a research intern at CVML Lab, advised by Prof. Angela Yao in NUS and Prof. Junbin Xiao in USTC.

Video generation & understandingWorld modelEmbodied AILong-term memory
The University of Hong Kong HK
Email Google Scholar GitHub ORCID OpenReview Bluesky

Research portfolio

Selected Projects

Selected publications, research systems, and ongoing experiments.

Recursive Game Creator workflow connecting Designer, Builder, Player, and Reviewer through iterative game development
2026 Preprint

Recursive Game Creator: An Agentic Product-Level Experience-Oriented Game Harness

Jiajun Chen , Haoyu Wu , Mingda Jia , Xihui Liu

arXiv

Paper ā†— Code ā†— Page ā†—
Overview of MuKV compressing a video stream into segment, frame, and patch-level KV caches
2026 CVPR Accepted Poster

MuKV: Multi-Grained KV Cache Compression for Long Streaming Video Question Answering

Junbin Xiao* , Jiajun Chen* , Tianxiang Sun , Xun Yang , Angela Yao

Computer Vision and Pattern Recognition (CVPR)

* Equal contribution

Paper ā†— Code ā†—
Architecture of PROMISE with prompt attention and hierarchical contrastive learning
2026 AAAI Accepted Poster

PROMISE: Prompt-Attentive Hierarchical Contrastive Learning for Robust Cross-Modal Representation with Missing Modalities

Jiajun Chen , Sai Cheng , Yutao Yuan , Yirui Zhang , Haitao Yuan , Peng Peng , Yi Zhong

Association for the Advancement of Artificial Intelligence (AAAI)

First author

Paper ā†—