Does DP synth text transfer useful knowledge or just superficial style mimicking? Existing benchmarks: saturated Introducing ContinuousBench: a hard (curr methods fail at ε=100! ) & leakage-proof benchmark for DP synth text! Followup to our #ICML2024 best paper 1/n
@thegautamkamath
-
Don’t let AI steal your thinking and personality in talks
By
–
In the last 48h:
– Jr researcher asked me wheter to use AI in making talks
– Saw two talks, with AI {slop, enhanced} slides Collected my thoughts and wrote a post. Tl;dr: don't steal your own thinking, don't remove *you* from your talks. Also, give a &#@% about your talks. -
New generative model research uses low-rank Nyström approximation for faster training
By
–
Ali did some amazing work on the hottest new generative model: drifting models (from Mingyang Deng @Goodeat258 et al., out of Kaiming He's group).
— Gautam Kamath (@thegautamkamath) 14 mai 2026
Speeds up training a lot using low rank Nyström approximation. Check out Ali's full thread. Paper and code available! https://t.co/MDSo47g4rUAli did some amazing work on the hottest new generative model: drifting models (from Mingyang Deng @Goodeat258 et al., out of Kaiming He's group). Speeds up training a lot using low rank Nyström approximation. Check out Ali's full thread. Paper and code available!
-
Accepted papers for COLT 2026 announced
By
–
Accepted papers for #COLT2026: https://
learningtheory.org/colt2026/accep
ted.html
… -
Research Incentives: Quality Over Quantity in Academia
By
–
Is that unfortunate? I feel like it sets the incentives as they should be: to try to make deep contributions rather than spamming papers.
-
AI Transparency: Validating and Disclosing AI-Generated Content
By
–
My new policy: if someone asks me to read something, I ask them how they used AI in creating it, and what validation/processing they applied to the AI outputs. (I disclose the same.) Been burned by giving too much attention to (undisclosed) slop folks have sent me…
-
Q&A with UW Cheriton CS colleague on algorithms
By
–
A Q&A with my new @UWCheritonCS colleague @Santho_Velusamy
, who works on algorithms and complexity, particularly streaming and other sublinear settings. Check it out! (Credits to comms guy Joe Petrik. Great articles and also great photos!) -
Learning Theorist Awarded Recognition for Research Contributions
By
–
Thanks to all my collaborators, my group @TheSalonML
, my academic communities (algorithmic stats, differential privacy, learning theory), my mentors, and the award committee. I appreciate the committee for being open-minded and recognizing two learning theorists this year! 6/7 -
Gautam Kamath Receives EATCS Presburger Award 2026
By
–
Laudatio: https://
eatcs.org/index.php/comp
onent/content/article/1-news/3034-presburger-award-2026-laudatio
… Article by our amazing comms guy @UWCheritonCS
, Joe Petrik: https://
cs.uwaterloo.ca/news/gautam-ka
math-receives-eatcs-presburger-award
… Before anyone says it, yes, I forgot to mention in the pic the estimator is computationally efficient. That's kinda the point! CYA at ICALP in London! 7/7 -
Joining NYU’s Leading AI Theory Research Group This Fall
By
–
The same will be true of NYU, when I join this Fall, as Aaron Bernstein also received it in 2023. In case you haven't been paying attention, NYU's theory group has become incredibly stacked the last couple years, and I am comparably intimidated and inspired by folks there. 4/7