Generative AI · Hardware-Aware Efficient AI

Jiankun Wang

I am an undergraduate student in Computer Science at Beihang University, expected to graduate in Fall 2027. My research interests center on generative AI and hardware-aware efficient AI algorithms.

Portrait of Jiankun Wang

Research Interests

Generative intelligence and efficient AI.

Generative AI

Generative models for language, vision, and video, with an emphasis on multimodal generation.

Hardware-Aware Efficient AI Algorithms

Designing efficient AI algorithms that account for hardware capabilities and constraints to reduce computational cost and improve practical performance.

Publications

Selected work

GitHub Profile

2026

Does More Retrieved Evidence Help Visual Retrieval-Augmented Generation with Diffusion Language Models?

Jiankun Wang, Yisen Gao, Ziwei Zhang, Xingcheng Fu, Jiaxin Bai, Chen Gao

arXiv preprint, 2026.

2026

Training-Free Sparse Attention for Fast Video Generation via Offline Layer-Wise Sparsity Profiling and Online Bidirectional Co-Clustering

Jiayi Luo, Jiayu Chen, Jiankun Wang, Cong Wang, Hanxin Zhu, Qingyun Sun, Chen Gao, Zhibo Chen, Jianxin Li

ICML 2026 poster.

2026

EmbodiedVAE: Disentangled Video VAE for Efficient and Controllable Embodied Manipulation

Jiayi Luo, Hanxin Zhu, Chen Gao, Jiankun Wang, Cong Wang, Tianyu He, Jianxin Li, Zhibo Chen

ECCV 2026.

Contact

Jiankun Wang 王健坤

Writing

One Thousand and One Nights

My existing MkDocs blog remains a separate reading space for notes, coursework, essays, and technical writing.

Enter the Blog