HE Xin

Research Scientist, CFAR, A*STAR, Singapore

ggbo_xiaoxin2.png

hexin.research@gmail.com

I am currently a Research Scientist at A*STAR CFAR, Singapore, working with Prof. Ong Yew Soon and Prof. Ivor W. Tsang. Prior to this, I completed my Ph.D. in Computer Science at Hong Kong Baptist University (HKBU), advised by Prof. Chu Xiaowen. I earned my Bachelor’s degree from the School of Electronic Information and Communications at Huazhong University of Science & Technology (HUST).

My research focuses on making large-scale AI models efficient, deployable, and trustworthy:

  • LLM Inference Optimization — Sparse MoE routing, speculative decoding, KV cache compression, token scheduling
  • Intellectual Property Protection — Model fingerprinting and IP attribution for LLMs at cluster scale
  • Neural Architecture Search (NAS) & AutoML — One-shot NAS, multi-objective search, resource-aware optimization

Highlights

news

Jun 18, 2026 📚 我的新书《动手学 AutoML:从 NAS 到大模型优化实战》正式出版啦! / My new book “Hands-On AutoML: From NAS to LLM Optimization in Practice” is now published!
Mar 06, 2026 Our paper “Lang-PINN: From Language to Physics-Informed Neural Networks via a Multi-Agent Framework” is accepted as Spotlight 🔥 to ICLR 2026 (Workshop on AI with Recursive Self-Improvement)!
Feb 24, 2026 Our paper “ExpertFlow: Efficient Mixture-of-Experts Inference via Predictive Expert Caching and Token Scheduling” is accepted to DAC2026!
Jan 26, 2026 Our paper “Ghost in the cloud: Your geo-distributed large language models training is easily manipulated” is accepted to ICLR2026!
May 16, 2025 Our paper “BurstGPT: A Real-world Workload Dataset to Optimize LLM Serving Systems” is accepted to KDD2025!
Feb 14, 2025 I am extremely honored to be one of the 31 recipients (selected from 194 applicants) of the A*STAR Career Development Fund (CDF)! news
Feb 23, 2024 I am honored to serve as a reviewer for ECCV’24.
Sep 03, 2023 Our paper “MedPipe: End-to-End Joint Search of Data Augmentation Policy and Neural Architecture for 3D Medical Image Classification” is accepted to IEEE MedAI2023!

latest posts

selected publications

  1. KDD
    BurstGPT: A Real-world Workload Dataset to Optimize LLM Serving Systems
    首个真实 LLM 服务突发请求数据集,揭示了现有调度策略在突发流量下的性能瓶颈,KDD 2025。 | First real-world LLM serving burst-request dataset, revealing scheduling bottlenecks under traffic spikes; KDD 2025.
    Yuxin Wang , Yuhan Chen , Zeyu Li , and 8 more authors
    2025
  2. AAAI
    NAS-LID: Efficient Neural Architecture Search with Local Intrinsic Dimension
    利用局部本征维度(LID)量化子网几何特性,以更低内存开销实现准确的超网分区,AAAI 2023 CCF-A。 | Uses local intrinsic dimension to characterize subnet geometry for accurate supernet partition at lower memory cost; AAAI 2023.
    Xin He , Jiangchao Yao , Yuxin Wang , and 5 more authors
    AAAI, 2023
  3. KBS
    AutoML: A survey of the state-of-the-art
    AutoML 领域最全综述,覆盖 NAS、HPO、数据增强等方向,KBS 2021,引用量超 2200。 | Comprehensive AutoML survey covering NAS, HPO, and data augmentation; KBS 2021, 2,700+ citations.
    Xin He , Kaiyong Zhao , and Xiaowen Chu
    Knowledge-Based Systems, 2021
  4. AAAI
    Automated Model Design and Benchmarking of 3D Deep Learning Models for COVID-19 Detection with Chest CT Scans
    首次将神经架构搜索应用于胸部 CT 新冠肺炎检测,自动设计三维深度学习模型,AAAI 2021 CCF-A。 | First NAS application to COVID-19 chest CT detection, automating 3D deep learning model design; AAAI 2021.
    Xin He , Shihao Wang , Xiaowen Chu , and 6 more authors
    In AAAI , 2021
  5. Book
    动手学 AutoML:从 NAS 到大语言模型优化实战
    国内首本系统覆盖 NAS、HPO 到大模型推理优化的 AutoML 实战书籍,机械工业出版社,2026。 | First Chinese textbook systematically covering NAS, HPO, and LLM optimization in practice; China Machine Press, 2026.
    Xin He , Xiaowen Chu , Kaiyong Zhao , and 3 more authors
    Jun 2026
    贺鑫(新加坡科技研究局 A*STAR / 香港浸会大学 HKBU),褚晓文(香港科技大学广州校区 HKUST(GZ)),赵开勇(其域创新有限公司 Qiyu Innovation Ltd.),王强(哈尔滨工业大学(深圳)HIT Shenzhen),唐桢桁(香港科技大学 HKUST),董佩杰(香港科技大学广州校区 HKUST(GZ))