I am a Ph.D. candidate in EDIC Computer and Communication Sciences at EPFL, advised by Prof. Sanidhya Kashyap. I received my B.S. in Computer Science with summa cum laude from Turing Class at Peking University. I also minored in Finance.

My research focuses on managing state across the memory hierarchy, such as pages, KV caches, and agent memory, from the OS kernel up to LLM agents. Every layer faces the same trade-off between fetching state and regenerating it, but the right balance differs across layers and shifts with the hardware and workload. I tackle it from two sides:

  • Scalable mechanisms that move pages through the hierarchy without becoming a bottleneck.
  • Learned policies that decide what to keep, fetch, or regenerate for the workload at hand.

Along the way, I have built systems at every level of the stack, from memory management in the Linux kernel to KV cache management for LLM inference and communication for LLM pretraining. I contribute regularly to open-source projects such as SGLang and Linux (both “kernels”, in a sense). I am also a member of the Swiss Apertus LLM project, where I contribute to pretraining mostly through UCCL.

Education

École Polytechnique Fédérale de Lausanne (EPFL), Switzerland

Sep. 2022 - Present

Ph.D. in Computer Science

Advised by Prof. Sanidhya Kashyap

Peking University, China

Sep. 2018 - Jul. 2022

B.S. in Computer Science; Minor in Finance

Advised by Prof. Chenren Xu

GPA: 3.83/4.00, Rank: 12/188

Work Experience

Rednote Rednote icon , Hangzhou, China
Jul. 2026 -
Research Intern, Dots
Mentors: Qianchao Zhu
Working on training-aware position independent KV cache and pretraining framework.
Meta Meta icon , London, UK
Jun. 2025 - Sep. 2025
Software Development Engineer Intern, Kernel Team
Working on memory allocation profiling and DAMON
Alibaba, DAMO Academy Alibaba, DAMO Academy icon , Hangzhou, China
Apr. 2022 - Aug. 2022
Research Intern, Networking Group
Mentors: Rui Miao, Junchen Guo, and Hao Nie
Worked on NFS optimization and 5G virtualization
EPFL, DSLab EPFL, DSLab icon , Lausanne, Switzerland
Oct. 2021 - Jan. 2022
Research Intern
Mentors: Lei Yan and George Candea
Worked on NFOS

Publications

CacheWeave: Learning to Recompute for Efficient Position-Independent KV Caching
Yueyang Pan , Qianchao Zhu , Yuliang Liu , Teng Ma , Sanidhya Kashyap
arXiv, 2026
[bib]
@misc{pan2026cacheweave,
  title = {CacheWeave: Learning to Recompute for Efficient Position-Independent KV Caching},
  author = {Pan, Yueyang and Zhu, Qianchao and Liu, Yuliang and Ma, Teng and Kashyap, Sanidhya},
  year = {2026},
  eprint = {2611.01735},
  archivePrefix = {arXiv}
}
NanoMem: Iterative Evidence Synthesis for Temporal-Causal Reasoning in Agent Memory
Yueyang Pan , Youliang Zhu , Nikhil Sarda , Haodong Zheng , Antoine Bosselut , Sanidhya Kashyap
Under submission to ICLR, 2027
Scalable Far Memory: Balancing Faults and Evictions Artifact AvailableArtifact FunctionalResults Reproduced
Yueyang Pan* , Yash Lala* , Musa Unal , Yujie Ren , Seung-seob Lee , Abhishek Bhattacharjee , Anurag Khandelwal , Sanidhya Kashyap
SOSP, 2025
[bib]
@inproceedings{pan2025sosp,
  title = {Scalable Far Memory: Balancing Faults and Evictions},
  author = {Pan, Yueyang and Lala, Yash and Unal, Musa and Ren, Yujie and Lee, Seung-seob and Shan, Yizhou and Bhattacharjee, Abhishek and Khandelwal, Anurag and Kashyap, Sanidhya},
  booktitle = {Proceedings of the ACM Symposium on Operating Systems Principles (SOSP)},
  year = {2025}
}
Tolerate It if You Cannot Reduce It: Handling Latency in Tiered Memory
Musa Unal , Vishal Gupta , Yueyang Pan , Yujie Ren , Sanidhya Kashyap
HotOS XX, 2025
[bib]
@inproceedings{unal2025hotos,
  title = {Tolerate It if You Cannot Reduce It: Handling Latency in Tiered Memory},
  author = {Unal, Musa and Gupta, Vishal and Pan, Yueyang and Ren, Yujie and Kashyap, Sanidhya},
  booktitle = {Proceedings of the ACM SIGOPS 20th Workshop on Hot Topics in Operating Systems (HotOS XX)},
  year = {2025}
}
Transparent Multicore Scaling of Single-Threaded Network Functions Artifact AvailableArtifact FunctionalResults Reproduced
Lei Yan , Yueyang Pan , Diyu Zhou , George Candea , Sanidhya Kashyap
EuroSys, 2024
[bib]
@inproceedings{yan2024eurosys,
  title = {Transparent Multicore Scaling of Single-Threaded Network Functions},
  author = {Yan, Lei and Pan, Yueyang and Zhou, Diyu and Candea, George and Kashyap, Sanidhya},
  booktitle = {Proceedings of the European Conference on Computer Systems (EuroSys)},
  year = {2024}
}
Monarch: A Fuzzing Framework for Distributed File Systems Artifact AvailableArtifact FunctionalResults Reproduced
Tao Lyu , Liyi Zhang , Zhiyao Feng , Yueyang Pan , Yujie Ren , Meng Xu , Mathias Payer , Sanidhya Kashyap
ATC, 2024
[bib]
@inproceedings{lyu2024atc,
  title = {Monarch: A Fuzzing Framework for Distributed File Systems},
  author = {Lyu, Tao and Zhang, Liyi and Feng, Zhiyao and Pan, Yueyang and Ren, Yujie and Xu, Meng and Payer, Mathias and Kashyap, Sanidhya},
  booktitle = {Proceedings of the USENIX Annual Technical Conference (ATC)},
  year = {2024}
}
Ship Your Critical Section, Not Your Data: Enabling Transparent Delegation with TCLOCKS Artifact AvailableArtifact FunctionalResults Reproduced
Vishal Gupta , Kumar Kartikeya Dwivedi , Yugesh Kothari , Yueyang Pan , Diyu Zhou , Sanidhya Kashyap
OSDI, 2023
[bib]
@inproceedings{gupta2023osdi,
  title = {Ship Your Critical Section, Not Your Data: Enabling Transparent Delegation with TCLOCKS},
  author = {Gupta, Vishal and Dwivedi, Kumar Kartikeya and Kothari, Yugesh and Pan, Yueyang and Zhou, Diyu and Kashyap, Sanidhya},
  booktitle = {Proceedings of the USENIX Symposium on Operating Systems Design and Implementation (OSDI)},
  year = {2023}
}
The First 5G-LTE Comparative Study in Extreme Mobility
Yueyang Pan* , Ruihan Li* , Chenren Xu
SIGMETRICS, 2022
[bib]
@inproceedings{pan2022sigmetrics,
  title = {The First 5G-LTE Comparative Study in Extreme Mobility},
  author = {Pan, Yueyang and Li, Ruihan and Xu, Chenren},
  booktitle = {ACM on Measurement and Analysis of Computing Systems (SIGMETRICS)},
  year = {2022}
}
Critique of “A parallel framework for constraint-based Bayesian network learning via Markov blanket discovery”
Jiaqi Si , Junyi Guo , Zhewen Hao , Wenyang He , Ruihan Li , Yueyang Pan , Zhenxin Fu , Chun Fan
TPDS, 2022
[bib]
@article{si2022tpds,
  title = {Critique of “A parallel framework for constraint-based Bayesian network learning via Markov blanket discovery”},
  author = {Si, Jiaqi and Guo, Junyi and Hao, Zhewen and He, Wenyang and Li, Ruihan and Pan, Yueyang and Fu, Zhenxin and Fan, Chun},
  journal = {IEEE Transactions on Parallel and Distributed Systems (TPDS)},
  year = {2022}
}
Critique of “MemXCT: Memory-Centric X-Ray CT Reconstruction With Massive Parallelization”
Zejia Fan , Yuchen Gu , Zhewen Hao , Yueyang Pan , Pengcheng Xu , Yuxuan Yan , Fangyuan Yang , Zhenxin Fu , Yun Liang
TPDS, 2021
[bib]
@article{fan2021tpds,
  title = {Critique of “MemXCT: Memory-Centric X-Ray CT Reconstruction With Massive Parallelization”},
  author = {Fan, Zejia and Gu, Yuchen and Hao, Zhewen and Pan, Yueyang and Xu, Pengcheng and Yan, Yuxuan and Yang, Fangyuan and Fu, Zhenxin and Liang, Yun},
  journal = {IEEE Transactions on Parallel and Distributed Systems (TPDS)},
  year = {2021}
}

Teaching

  • CS477: Advanced Operating Systems
    Fall 2025, TA, EPFL
  • CS300: Data-intensive Systems
    Spring 2025, TA, EPFL
  • CS477: Advanced Operating Systems
    Fall 2024, TA, EPFL
  • CS300: Data-intensive Systems
    Spring 2024, TA, EPFL
  • CS323: Introduction to Operating Systems
    Fall 2023, TA, EPFL
  • CS323: Introduction to Operating Systems
    Fall 2022, TA, EPFL
  • 04832250: Computer Networks
    Fall 2021, TA, Peking University
  • Turing Class: Summer Reading Group
    Summer 2019, Organizer, Peking University

Awards & Honors

  • Teaching Assistant Award, EDIC, EPFL, 2024
  • Outstanding Graduates of Beijing, Beijing, 2022
  • Outstanding Graduates of Peking University, Peking University, 2022
  • China National Scholarship, China, 2021
  • Merit Student, Peking University, 2021
  • ASC Student Supercomputer Challenge 20-21 — 4th Place, Worldwide, 2021
  • Second place, SC Student Cluster Competition 2020, Worldwide, 2020
  • John Hopcroft Scholarship, Turing Class, 2020
  • Schlumberger Scholarship, Peking University, 2019