Portrait of Huiqiang Jiang

Huiqiang Jiang

Research SDE 2

About

Huiqiang is a Research SDE in MSRA Shanghai Lab.

Huiqiang’s research primarily concentrates on efficient methods to accelerate inference or training, including dynamic sparse attention (MInference (opens in new tab), RetrievalAttention (opens in new tab)), prompt compression (LLMLingua (opens in new tab)), KV-cache compression, speculative decoding, model compression, sparse inference (PIT (opens in new tab)), neural architecture search (NAS), and efficient tuning, with a particular emphasis on LLMs. Additionally, he is interested in addressing typical challenges in natural language processing, such as information extraction.

He’s looking for one research intern in efficient methods. Please get in touch with him (hjiang[aT]microsoft[DoT.]com) if you are interested in the research topics.