Hi! 😄 I’m Xinyi Liu, a fourth-year Ph.D. student in the School of Computer Science at Peking University and a member of the DAIR Lab, advised by Professor Bin Cui. I received my B.S. in Cyberspace Security from the School of Computer Science and Technology at Harbin Institute of Technology, Weihai in June 2023. I am a lead developer of Hetu-Galvatron, an open-source system for automated distributed training of Transformer models.

My research focuses on deep learning systems, particularly efficient infrastructure for foundation-model pre-training and post-training.

I am actively seeking collaborations and open to discussions on related topics. Feel free to reach out!

📖 Education

🔥 News

[2026.07] - I was selected as one of 44 recipients of the 2026 CIE–Tencent Ph.D. Student Research Incentive Program (Hunyuan Foundation Model Track)! 🎉
[2026.04] - Our paper DARTS has been accepted to ICML 2026! 🎉
[2026.01] - Our paper LAER-MoE has been accepted to ASPLOS 2026! 🎉
[2025.10] - Our paper Galvatron-2 has been accepted to FAISys 2025! 🎉
[2025.06] - I gave a talk on Galvatron at PyTorch Day China, Beijing, China. ⭐️
[2025.05] - I gave a talk on Galvatron at the Kunpeng Ascend Developer Conference (KADC), Beijing, China. ⭐️
[2025.01] - Our paper FlexSP has been accepted to ASPLOS 2025! 🎉
[2025.01] - Our paper NetMoE has been accepted to ICLR 2025 as a Spotlight! 🎉

[ICML] Yujie Wang, Siwei Chen, Longzan Luo, Xinyi Liu, Xupeng Miao, Fangcheng Fu, Bin Cui. “DARTS: Distribution-Aware Active Rollout Trajectory Shaping for Accelerating LLM Reinforcement Learning.”
[ASPLOS] Xinyi Liu, Yujie Wang, Fangcheng Fu, Xuefeng Xiao, Huixia Li, Jiashi Li, Bin Cui. “LAER-MoE: Load-Adaptive Expert Re-layout for Efficient Mixture-of-Experts Training.”

[arXiv] Xinyi Liu, Yujie Wang, Shenhan Zhu, Fangcheng Fu, Qingshuo Liu, Guangming Lin, Bin Cui. “Galvatron: An Automatic Distributed System for Efficient Foundation Model Training.”
[ICLR Spotlight] Xinyi Liu, Yujie Wang, Fangcheng Fu, Xupeng Miao, Shenhan Zhu, Xiaonan Nie, Bin Cui. “NetMoE: Accelerating MoE Training through Dynamic Sample Placement.”
[ASPLOS] Yujie Wang, Shiju Wang, Shenhan Zhu, Fangcheng Fu, Xinyi Liu, Xuefeng Xiao, Huixia Li, Jiashi Li, Faming Wu, Bin Cui. “FlexSP: Accelerating Large Language Model Training via Flexible Sequence Parallelism.”

Lead Developer | GitHub | Documentation

A high-performance automatic distributed training system for Transformer models and LLMs, independently developed and open-sourced by PKU-DAIR Lab.

Automatic Parallelism Optimization: Efficiently search for optimal strategies via cost modeling
Fine-grained Hybrid Parallelism: Layer-wise flexible configuration (DP, SDP/ZeRO, PP, TP, SP, CKPT)
Workload Versatility: BERT, GPT, T5, LLaMA, Vision Transformers, multi-modality models

CIE–Tencent Ph.D. Student Research Incentive Program (Hunyuan Foundation Model Track), one of 44 recipients (2026)
ASPLOS 2026 Student Travel Grant (2026)
Kunpeng Ascend Outstanding Young Contributor in Scientific Research and Innovation (2025)
National Scholarship (2022)
Candidate for the 13th China Youth Science and Technology Innovation Award (2022)
7th Place (Top 0.2%), 26th CCF Certified Software Professional Certification (2022)
Runner-up, Gold Medal, 12th Shandong Provincial ICPC Collegiate Programming Contest (2022)
Outstanding Student of Higher Education Institutions (2021)
Gold Medal, 2021 China Collegiate Programming Contest, Harbin Site (2021)
Silver Medal, 46th ICPC Asia Regional Contest, Jinan Site (2021)

Galvatron: An Automatic Distributed System for Efficient Large-Scale Transformer Training. PyTorch Day China, Beijing, China, June 2025
Galvatron: An Automatic Distributed System for Efficient Large-Scale Transformer Training. Kunpeng Ascend Developer Conference (KADC), Beijing, China, May 2025