Provable algorithms for inference, optimization, and quantum computation.
May 2026 — Started as Assistant Professor at HKUST (Guangzhou)
Research
Three core areas
Structured inference
Bethe methods, graph covers, and message passing for counting and inference on graphical models.
Provable optimization
Permanent bounds and algorithmic guarantees for hard combinatorial problems.
Quantum & tensors
Tensor networks and distributed quantum systems for scalable computation.
Selected work
Representative papers
Graphical models and Bethe methods
Graph covers and Bethe approximation for inference and counting — our longest-running thread.
Tensor methods
Tensor-network methods for high-dimensional computation.
Efficient LLMs
Provably linear-time attention and post-training pruning through block reconstruction and Gauss–Seidel refinement.
Latest
What's new
Sep 2026
NeurIPS 2026, Dec. 2026 · Zhuohua Li, Maoli Liu, Yuwen Huang, Cheng Wen, Jie Su, Cong Tian, Shengchao Qin, and John C.S. Lui
Sep 2026
Block-OBS-GS: Exact Per-Block Joint Brain Surgery with Gauss–Seidel Refinement for LLM Pruning To appear
NeurIPS 2026, Dec. 2026 · Yuwen Huang and Xiang Pan
Aug 2026
Major revision resubmitted to IEEE Transactions on Information Theory
Jul 2026
Two research grants awarded as Principal Investigator New
Guangzhou Municipal Education Bureau · Department of Education of Guangdong Province
May 2026
Started as Assistant Professor at HKUST (GZ)
DSA Thrust, Information Hub
May 2026
ICML 2026
Mar 2026
Submitted to Quantum
Team
Research assistants
Members of the group since 2026.
PhD graduate, CSE, CUHK
PhD student (Year 2), ECE, UBC
PhD student (Year 2), CSE, CUHK
Yu Jiang
PhD student (Year 1), CSE, CUHK
Deyun Zhang
Master's graduate, Mathematics, SUSTech
Yingxian Hui
Undergraduate, Mathematics, CUHK (Shenzhen)
Open positions
Join the group
Building a team at HKUST (Guangzhou).
Full tuition and competitive stipend
Mentorship targeting IEEE TIT, ISIT, ICML
Greater Bay Area research ecosystem
Small group, close advising
RA PhD Postdoc