Xingyang Li

        [CV]


Contact:

brucelee_sjtu (AT) sjtu (dot) edu (dot) cn
xyli05 (AT) mit (dot) edu

I am a fourth-year undergraduate student at ACM Honors Class, Shanghai Jiao Tong University. I work on efficient algorithms and systems for deep learning applications, particularly generative AI.

Now I am a visiting student at MIT HAN Lab advised by Prof. Song Han. During my junior year, I am very fortunate to work with Prof. Yu Feng and Prof. Guohao Dai.

💡 Let's see our recent work: [NeurIPS'25] Radial Attention, featuring O(nlogn) Sparse Attention for training/inference acceleration of video diffusion models! Feel free to try it with the official implementation or on the state-of-the-art WanVideo inference engine.

News [Show All]

Education

Massachusetts Institute of Technology, USA
Student Research Intern • Jul. 2025 - Present
Advisor: Prof. Song Han

Shanghai Jiao Tong University, China
B.Eng. in Computer Science • Sept. 2022 - June 2026 (Expected)
Member of ACM Honors Class (Top 5% students with interest in research)
GPA: 4.03/4.3, Rank: 3/30
Advisor: Prof. Yong Yu

Selected Publications

* indicates equal contribution


Radial Attention
Radial Attention: O(nlogn) Sparse Attention with Energy Decay for Long Video Generation
Xingyang Li*, Muyang Li*, Tianle Cai, Haocheng Xi, Shuo Yang, Yujun Lin, Lvmin Zhang, Songlin Yang, Jinbo Hu, Kelly Peng, Maneesh Agrawala, Ion Stoica, Kurt Keutzer, Song Han
NeurIPS 2025 / arXiv / Code (500+ stars) / Website / Blog
Nebula
Nebula: Enable City-Scale 3D Gaussian Splatting in Virtual Reality via Collaborative Rendering and Accelerated Stereo Rasterization
He Zhu, Zheng Liu, Xingyang Li, Anbang Wu, Jieru Zhao, Fangxin Liu, Yiming Gan, Jingwen Leng, Yu Feng
ASPLOS 2026 (Oral Presentation) / arXiv
SLTarch
SLTarch: Towards Scalable Point-Based Neural Rendering by Taming Workload Imbalance and Memory Irregularity
Xingyang Li*, Jie Jiang*, Yu Feng, Yiming Gan, Jieru Zhao, Zihan Liu, Jingwen Leng, Minyi Guo
ICCAD 2025 (Oral Presentation) / arXiv
Harnessing Conventional Video Processing
Harnessing Conventional Video Processing Insights for Emerging 3D Video Generation Models: A Comprehensive Attention-aware Way
Tianlang Zhao*, Jun Liu*, Xingyang Li*, Li Ding, Jinhao Li, Shuaiheng Li, Jinbo Hu, Guohao Dai
DAC 2025 (Oral Presentation) / paper

Teaching Experience

  • Teaching Assistant for Machine Learning (02/2025 – 06/2025)
  • Teaching Assistant for Mathematical Logic (02/2024 – 06/2024)
  • Teaching Assistant for Programming (09/2023 – 02/2024)

Role: giving lectures and recitation classes, writing documents and sample solutions, grading homework, creating exam questions, designing ML-related final project lists

Honors & Awards

  • 2025 SenseTime Scholarship (30 winners nationwide with aspiration in AI research)
  • 2022-2025 Zhiyuan Honorary Scholarship
  • 2023-2025 Academic Excellence Scholarship (Second Prize)
  • 2023-2025 Commercial Sponsorship Scholarship (Second Prize)

Miscellaneous