Wooseok Gwak

Hi, I am a first-year PhD student in Computer Science at Georgia Tech, advised by Prof. Anand Iyer in the NEXS group. Before joining Georgia Tech, I received my B.S. in Computer Science with a minor in Mathematics from KAIST (Summa Cum Laude), where I was a research intern at the CASYS lab advised by Prof. Youngjin Kwon.

My research interests lie in computer systems for machine learning. I focus on building efficient AI infrastructure, including LLM serving systems, memory and KV cache management, and the runtime mechanisms that make them fast and dependable.

If you’re interested in discussing research or exploring collaboration opportunities, I’d love to connect. Please feel free to reach out to me at wgwak3@gatech.edu.

selected publications

Preprint

S^3: Unlocking the Full Potential of Sparse Attention for Long Context LLM Serving

Wooseok Gwak^*, Sangjin Choi^*, Sukmin Cho, Jin Heo, Jinho Choi, and Youngjin Kwon

Preprint, 2025

Bib

@article{Gwak2025S3,
  title = {S^3: Unlocking the Full Potential of Sparse Attention for Long Context LLM Serving},
  author = {Gwak, Wooseok and Choi, Sangjin and Cho, Sukmin and Heo, Jin and Choi, Jinho and Kwon, Youngjin},
  journal = {Preprint},
  year = {2025},
}

USENIX Security’24

BUDAlloc: Defeating Use-After-Free Bugs by Decoupling Virtual Address Management from Kernel

Junho Ahn, Jaehyun Lee, Kanghyuk Lee, Wooseok Gwak, Minseong Hwang, and Youngjin Kwon

In Proceedings of USENIX Security, 2024

Bib PDF Code Slides

@article{Ahn2024BudAlloc,
  title = {BUDAlloc: Defeating Use-After-Free Bugs by Decoupling Virtual Address Management from Kernel},
  author = {Ahn, Junho and Lee, Jaehyun and Lee, Kanghyuk and Gwak, Wooseok and Hwang, Minseong and Kwon, Youngjin},
  journal = {In Proceedings of USENIX Security},
  year = {2024},
}