Chenyu He何宸禹

About

My research interest lies in LLM serving systems, particularly in the memory hierarchy for long-context inference, and more broadly in serverless model loading and multi-agent systems for AIOps. I am currently advised by Dingyu Yang and Huan Li.

News

• 2026.07 R/W Cache published at ACM TURC 2026 (Hangzhou), and awarded the ByteDance Excellent Poster Award at the Turing Student Forum.
• 2026.07 Summer research intern at 先进计算与智能工程国家级重点实验室 (Wuxi), working on large-model deployment.
• 2026.05 Token Economics for LLM Agents posted to arXiv (arXiv:2605.09104).
• 2026.01 Interned at the Big Data Bureau of Kecheng District, Quzhou, working on asset inventory and vulnerability triage for government systems.

Publications

Published

1. R/W Cache: Read/Write-Aware KV-Cache Tiering for LLM Serving with CXL Memory
   Chenyu He, Dingyu Yang
   ACM TURC '26 — Proceedings of the ACM Turing Award Celebration Conference, China · pp. 165–167
   DOI  /  BibTeX

Preprints

1. Token Economics for LLM Agents: A Dual-View Study from Computing and Economics
   Yuxi Chen, Junming Chen, Chenyu He, Yiwei Li, Yicheng Ji, Yifan Wu, Dingyu Yang, Lansong Diao, Lidan Shou, Hongliang Zhang, Huan Li, Gang Chen
   arXiv:2605.09104, 2026
   arXiv

Awards

• 2026 ByteDance Excellent Poster Award, Turing Student Forum, ACM TURC 2026 — for R/W Cache.
• 2026 Gold Award, 18th “Dandelion” Innovation Competition, for intelligenergy (physics-informed neural network for multi-fuel flame radiation modeling); patent granted.
• 2026 Provincial Second Prize, China Undergraduate Mathematical Contest in Modeling (CUMCM).