About Me
I am a third-year PhD candidate in Data Science and Analytics at The Hong Kong University of Science and Technology (Guangzhou), advised by Prof. Zeyi Wen.
I work on machine learning systems for efficient long-context inference in large language models. My research brings together algorithm and system design, with a focus on KV cache optimization, low-precision GPU computing, and speculative decoding.
Publications
Channel-Aware Mixed-Precision Quantization for Efficient Long-Context Inference
Strong Drafts Need Compact Memories: Long-Context Speculative Decoding with Compressed KV Cache
DeInfer: Efficient Parallel Inferencing for Decomposed Large Language Models
Dovetail: Enabling FP4 KV Cache for Efficient LLM Serving on FP8-Capable GPUs
Education
09/2024 – present
Ph.D. in Data Science and Analytics
The Hong Kong University of Science and Technology (Guangzhou)
Guangzhou, China · Advised by Prof. Zeyi Wen
09/2018 – 06/2022
Teaching
Teaching assistant at HKUST(GZ).
DSAA 3032: Introduction to High-Performance and Parallel Computing, Summer 2026.
DSAA 2042: Computer Architecture and Systems, Fall 2025.