Me
Jiayu Chen
Peking University
jiayu.chen.25 [at] stu.pku.edu.cn

About

I am currently a Ph.D. student at Peking University, School of Computer Science, working with Prof. Xiang Chen in the IFLab. I received my B.Eng. degree in Software Engineering from the School of Information and Software Engineering, University of Electronic Science and Technology of China, in 2025.

Recently, I have been working on efficient video generation, with a focus on sparse attention and real-time generation for video generation models. If you are interested in discussing related topics, feel free to send me an email.

Interests

  • Efficient Video Generation
  • Video World Models
  • On-device AI Systems

Education

Peking University
  • Ph.D. in Computer Science and Technology, Sep. 2025 - Present
  • Advisor: Prof. Xiang Chen
University of Electronic Science and Technology of China
  • B.Eng. in Software Engineering, Sep. 2021 - Jun. 2025
  • Advisor: Prof. Yong Liao

News

  • 2026/07 Two papers on Adaptive Caching for Diffusion Transformer are accepted by ACM MM 2026!
  • 2026/06 One paper on Cloud-Edge Collaborative Video Generation is accepted by ECCV 2026!
  • 2026/04 One collaborative paper on Sparse Attention for Video Generation is accepted by ICML 2026!
  • 2026/02 One paper and two collaborative papers on Hardware-Software Co-Design are accepted by DAC 2026!
  • 2026/01 One paper on Token Pruning for VAR Models is accepted by ICLR 2026!
  • 2025/11 One collaborative paper on Quantization for MoE Models is accepted by DATE 2026!

Experiences

Jul, 2026 - Present
Research Intern @ JD Explore Academy
Advisor: Dr. Haoyang Huang
Sep, 2025 - Present
Ph.D. Student @ Peking University
Advisor: Prof. Xiang Chen
Jan, 2024 - Jul, 2024
Research Intern @ THU(AIR)
Advisor: Prof. Yunxin Liu, Prof. Yuanchun Li and Dr. Xiangyu Li
Sep, 2021 - Jun, 2025
Undergraduate Student @ UESTC
Advisor: Prof. Yong Liao

Publications ( / )

EcoFrame adaptive visual evidence scheduling overview
When and Where to Look: Adaptive Visual Evidence Scheduling for Efficient Long Video Understanding
Ke Li*, Jiayu Chen*, Maoliang Li, Zihao Zheng, Hailong Zou, Hengyi Zhang, Xuanzhe Liu, Xiang Chen.
Arxiv 2026
[Paper]
EchoCache caching method overview
EchoCache: Energy-Guided Cross-Modal Caching for Efficient Audio-Driven Video Generation
Jiayu Chen, Xiaoyu Wu, Rongshan Gao, Maoliang Li, Zihao Zheng, Xinhao Sun, Hailong Zou, Guojie Luo, Xiang Chen.
ACM MM 2026, CCF-A
[Paper] [Code]
MoECa framework overview
MoECa: Aligning Feature Reuse with Expert Decomposition in Diffusion Transformers
Maoliang Li, Haojing Chen, Jiayu Chen, Zihao Zheng, Xinhao Sun, Hailong Zou, Xiang Chen.
ACM MM 2026, CCF-A
[Paper] [Code]
Token Radius Attention overview
Token Radius Attention for Efficient Video Generation
Jiayu Chen*, Zhikun Jiang*, Maoliang Li, Jiayi Luo, Jiawei Yang, Zihao Zheng, Hengyi Zhang, Guojie Luo, Xiang Chen.
Arxiv 2026
[Paper] [Code]
EcoVideo overview
EcoVideo: Entropy-Orchestrated Video Generation Paradigm in Cloud-Edge Dynamics
Jiayu Chen*, Hengyi Zhang*, Maoliang Li, Minyu Li, Zihao Zheng, Xuanzhe Liu, Guojie Luo, Xiang Chen.
ECCV 2026, CCF-B
[Paper] [Code]
Publication image placeholder
Pyramid Forcing: Head-Aware Pyramid KV Cache Policy for High-Quality Long Video Generation
Jiayu Chen, Junbei Tang, Wenbiao Zhao, Maoliang Li, Jiayu Luo, Zihao Zheng, Jiawei Yang, Guojie Luo, Xiang Chen.
Arxiv 2026
[Paper] [Code] [Website]
Publication image placeholder
Attention Sparsity is Input-Stable: Training-Free Sparse Attention for Video Generation via Offline Sparsity Profiling and Online QK Co-Clustering
Jiayi Luo, Jiayu Chen, Jiankun Wang, Cong Wang, Hanxin Zhu, Qingyun Sun, Chen Gao, Zhibo Chen, Jianxin Li.
ICML 2026, CCF-A
[Paper] [Code]
Publication image placeholder
ToProVAR: Efficient Visual Autoregressive Modeling via Tri-Dimensional Entropy-Aware Semantic Analysis and Sparsity Optimization
Jiayu Chen, Ruoyu Lin, Zihao Zheng, Jingxin Li, Maoliang Li, Guojie Luo, Xiang Chen.
ICLR 2026, CCF-A
[Paper] [Code]
Publication image placeholder
HeRo: Adaptive Orchestration of Agentic RAG on Heterogeneous Mobile SoC
Maoliang Li*, Jiayu Chen*, Zihao Zheng, Ziqian Li, Xinhao Sun, Guojie Luo, Chenchen Liu, Xiang Chen.
DAC 2026, CCF-A
[Paper] [Code]
[Full List]

Selected Honors & Awards

Academic Services

Conference Reviewer:
  • Conference on Neural Information Processing Systems (NeurIPS)
Top