Scholar
Xiuyu Li
Google Scholar ID: Kq0feJIAAAAJ
UC Berkeley
Citations & Impact
Citations
1,997
H-index
18
i10-index
19
Publications
20
Co-authors
10
list available
Publications
Resume (English only)
Academic Achievements
- Published several papers, such as 'Learning Adaptive Parallel Reasoning with Language Models' (COLM, 2025), which introduced the APR framework to improve the reasoning capabilities of language models; other works include SparseLoRA (ICML'25), Q-Diffusion (ICCV'23), SVDQuant (ICLR'25), etc.
Research Experience
- Involved in multiple research projects covering efficient generative models (quantization & sparsity), long-context LLMs/VLMs, ML systems, etc.
Education
- Currently a Ph.D. candidate at UC Berkeley, affiliated with Berkeley AI Research (BAIR), advised by Prof. Kurt Keutzer. Received a B.A. in Computer Science and Math from Cornell University, where he worked with Profs. Zhiru Zhang, Vitaly Shmatikov, and Song Han.
Background
- Research interests include enhancing the reasoning capabilities of large language models and developing scalable AI agents. Also has broad expertise in making generative models more efficient in both training and inference across language and vision.
Miscellany
- Actively seeking full-time member of technical staff and AI researcher/engineer positions in the industry.
