diff --git a/_cite/.cache/cache.db b/_cite/.cache/cache.db index 9dfbb8a..f1c8359 100644 Binary files a/_cite/.cache/cache.db and b/_cite/.cache/cache.db differ diff --git a/_data/citations.yaml b/_data/citations.yaml index 4ef39cb..78faeb2 100644 --- a/_data/citations.yaml +++ b/_data/citations.yaml @@ -1,5 +1,26 @@ # DO NOT EDIT, GENERATED AUTOMATICALLY +- id: arxiv:2601.00397 + title: 'Revati: Transparent GPU-Free Time-Warp Emulation for LLM Serving' + authors: + - Amey Agrawal + - Mayank Yadav + - Sukrit Kumar + - Anirudha Agrawal + - Garv Ghai + - Souradeep Bera + - Elton Pinto + - Sirish Gambhira + - Mohammad Adain + - Kasra Sohrab + - Chus Antonanzas + - Alexey Tumanov + publisher: arXiv + date: '2026-01-05' + link: https://arxiv.org/abs/2601.00397 + image: images/publication_thumbnails/revati.png + plugin: sources.py + file: sources.yaml - id: arxiv:2507.09019 title: On Evaluating Performance of LLM Inference Serving Systems authors: @@ -27,15 +48,15 @@ - Po-An Tsai - Zhiding Yu - Alexey Tumanov - publisher: arXiv + publisher: 21st European Conference on Computer Systems, 2026, Edinburgh date: '2025-08-14' link: https://arxiv.org/abs/2502.14051 image: images/publication_thumbnails/rocketkv.png plugin: sources.py file: sources.yaml - id: arxiv:2409.17264 - title: 'Medha: Efficiently Serving Multi-Million Context Length LLM Inference Requests - Without Approximations' + title: 'No Request Left Behind: Tackling Heterogeneity in Long-Context LLM Inference + with Medha' authors: - Amey Agrawal - Haoran Qiu @@ -47,7 +68,7 @@ - Alexey Tumanov - Esha Choukse publisher: arXiv - date: '2025-06-23' + date: '2025-11-27' link: https://arxiv.org/abs/2409.17264 image: images/publication_thumbnails/medha.png plugin: sources.py diff --git a/_data/sources.yaml b/_data/sources.yaml index 4df2ec9..95e7c77 100644 --- a/_data/sources.yaml +++ b/_data/sources.yaml @@ -1,7 +1,10 @@ +- id: arxiv:2601.00397 # revati + image: images/publication_thumbnails/revati.png - id: arxiv:2507.09019 # on evaluating llm inf.. image: images/publication_thumbnails/eval-checklist.png - id: arxiv:2502.14051 # maya image: images/publication_thumbnails/maya.png + publisher: "21st European Conference on Computer Systems, 2026, Edinburgh" - id: arxiv:2502.14051 # rocketkv image: images/publication_thumbnails/rocketkv.png - id: arxiv:2409.17264 # medha diff --git a/_members/angelina-zhou b/_members/angelina-zhou deleted file mode 100644 index 5481aa6..0000000 --- a/_members/angelina-zhou +++ /dev/null @@ -1,7 +0,0 @@ ---- -name: Angelina Zhou -image: https://media.licdn.com/dms/image/C5603AQExmk75bxmQNw/profile-displayphoto-shrink_400_400/0/1589406551907?e=1715817600&v=beta&t=TJMjTpUKFHZG44sQG8GVoF4AQrgHXx16sgoao0e3WAU -role: undergrad -links: - home-page: https://www.linkedin.com/in/angelina-zhou-az ---- \ No newline at end of file diff --git a/_members/chathurvedhi-talapaneni.md b/_members/chathurvedhi-talapaneni.md deleted file mode 100644 index 24c6358..0000000 --- a/_members/chathurvedhi-talapaneni.md +++ /dev/null @@ -1,8 +0,0 @@ ---- -name: Chathurvedhi Talapaneni -image: https://cheata-77.github.io/assets/profile.jpeg -role: masters -links: - home-page: www.linkedin.com/in/chathurvedhi ---- -I’m a software systems enthusiast with interests in Low-Level Programming, Computer Networks, and Inference Engines. I’m currently pursuing my Master’s in Computer Science at the Georgia Institute of Technology, Atlanta, and earned my Bachelor’s in Computer Science and Engineering from the Indian Institute of Technology, Madras, in 2024. \ No newline at end of file diff --git a/_members/faisal-baig.md b/_members/faisal-baig.md deleted file mode 100644 index fa26a7d..0000000 --- a/_members/faisal-baig.md +++ /dev/null @@ -1,9 +0,0 @@ ---- -name: Faisal Baig -image: https://faisab.github.io/assets/img/pfp2.jpg -role: masters -links: - home-page: https://www.linkedin.com/in/baig-faisal/ ---- - -Hey! I’m an MS CS student at Georgia Tech, where I also completed my undergrad. Currently interested in systems research for machine learning and building infrastructure for inference and training. diff --git a/_members/nandan-parikh.md b/_members/nandan-parikh.md deleted file mode 100644 index 2ffc3ec..0000000 --- a/_members/nandan-parikh.md +++ /dev/null @@ -1,10 +0,0 @@ ---- -name: Nandan Parikh -image: https://nan-done.github.io/images/with_chandru.jpg -role: masters -links: - github: lelouch0204 - home-page: https://nan-done.github.io ---- - -I am a first year MS CS student at Georgia Tech planning to specialize in ML. Most recently I was working at Flipkart as an SDE-2 on their travel search systems. During the course of my Master's degree I'll focus on better understanding Systems for ML and also building on my experience in Computer Vision. I graduated from BITS Pilani, India with a B.E. in CS. \ No newline at end of file diff --git a/_members/uday-goyat.md b/_members/uday-goyat.md deleted file mode 100644 index 8a647d7..0000000 --- a/_members/uday-goyat.md +++ /dev/null @@ -1,9 +0,0 @@ ---- -name: Uday Goyat -image: https://udaygoyat45.github.io/assets/portrait.png -role: masters -links: - home-page: https://udaygoyat45.github.io ---- - -Hey, this is Uday Goyat. I’m a Master’s student in Computer Science at Georgia Tech, where I focus on systems and machine learning. I’m especially excited about designing scalable infrastructure that helps bring cutting-edge AI research into real-world use. \ No newline at end of file diff --git a/images/publication_thumbnails/revati.png b/images/publication_thumbnails/revati.png new file mode 100644 index 0000000..0355575 Binary files /dev/null and b/images/publication_thumbnails/revati.png differ diff --git a/index.md b/index.md index 5816694..b139e67 100644 --- a/index.md +++ b/index.md @@ -5,6 +5,7 @@ The System for AI Lab (SAIL) at Georgia Tech, led by Prof. Alexey Tumanov, speci # Recent News +- Our paper on LLM training performance modeling via GPU emulation, [Maya](https://arxiv.org/abs/2503.20191) has been accepted at EuroSys'26. - 🎉 Congratulations to Payman Behnam, Amey Agrawal, Alind Khare, and Dhruv Garg! Three papers accepted at ACM SIGOPS Operating Systems Review, July 2025. - We are looking for contributors for our new inference engine [Vajra](https://project-vajra.github.io/). ⚡️ - Our papers on common anti-patterns in LLm Inference systems evaluations is now on [Arxiv](https://arxiv.org/pdf/2507.09019).