Sahil Joshi
I am a second-year Ph.D. student advised by Dr. Anshumali Shrivastava with a research focus on efficient training/inference of large language models and their applications. I hold a Bachelor of Science in Computer Science from the University of Massachusetts, Amherst. Prior to starting my Ph.D., I gained two years of industry experience as a Software Engineer at Dell Technologies. Beyond work, I’m passionate about travelling around the world, getting immersed in gripping thriller/horror shows, and staying active by playing table tennis competitively.
News
- I recently completed my third semester and have finished 4 of the 6 required courses in my Ph.D. program. Our recent work, “Scout Before You Attend: Sketch-and-Walk Sparse Attention for Efficient LLM Inference” was accepted to ICML 2026!
- My first research paper, “RACE Attention: A Strictly Linear-Time Attention Layer for Training on Outrageously Large Contexts” was accepted to ICLR 2026!
Awards & Service
- I will be serving as a Reviewer for NeurIPS 2026!
- I have been re-elected as the President of the Computer Science Graduate Student Association (CSGSA) for the 2026–2027 academic year.
- In 2023, I graduated from University of Massachusetts, Amherst, where I was honored with the Outstanding Undergraduate Achievement Award for my academic and research contributions.
