🧑‍🔬 Chengruidong Zhang

Affiliations

• Microsoft (United States) · Redmond, United States — current · 2024–2026 on 8 papers
• Microsoft Research Asia (China) · Beijing, China — current · 2026 on 2 papers
• Alibaba Group (China) · Hangzhou, China — current · 2026 on 1 paper

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 2 more institutions on single bylines are not listed.

Topics

Language models & pretraining2

Papers

The most cited first (OpenAlex and our library).

1. MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention (2024 · 197 citations)
   Long-context LLMs face a serious delay before they produce their first token. During pre-filling, attention cost grows quadratically with prompt length. An 8B model needs about half an hour on one A100 to ingest a 1M-token prompt. MInference is a training-free way to cut this delay. The authors observe that attention…
2. LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens (2024 · 138 citations)
   Most large language models can only work with a fixed span of text. LLaMA2, for example, is limited to 4096 tokens, and earlier extension methods stopped around 128k. Going further is hard for three reasons: unseen position indices destabilize training, very long training texts are scarce and expensive to use, and…

Also in our library

1 more work of theirs is in our library without a page here yet; it will appear as we publish them.

Co-authors here

12 co-authors with a page, the most papers together first.

• Chin-Yew Lin
• Dongsheng Li
• Fan Yang
• Huiqiang Jiang
• Li Lyna Zhang
• Lili Qiu
• Mao Yang
• Ning Shang
• Qianhui Wu
• Xufang Luo
• Yucheng Li
• Yuqing Yang

Cited by

Papers of our library cite their works 347 times (of the 140 884 papers whose bibliographies we read); 12 citing papers with a page here →.

This page in Markdown → https://md.research.elamur.ai/free/en/authors/chengruidong-zhang/