AI monitoring is key to responsible deployment. Our #ICML2025 paper develops approaches for 3 main goals:
1) *Adapting* to mild data shifts
2) *Quickly Detecting* harmful shifts
3) *Diagnosing* cause of degradation
🧵w/ @xinghan0@anqi_liu33@suchisaria
https://t.co/IZ04bMyVl7
I've just arrived in Vilnius, Lithuania for @ECMLPKDD and I’m happy to present our paper, "Novel Node Category Detection Under Subpopulation Shift”
Paper link: https://t.co/mud47OL1Hy
Had a (positive) “deer-in-the-headlights” moment today when scrolled to these kind & generous words from @predict_addict on our new paper w/ @samuel_stanton_@anqi_liu33@suchisaria! 😆https://t.co/9yv95OatsY
Humbled & grateful to share this work. Full paper summary thread soon!
We're thrilled to release our Hugging Face demo for the one-step InstaFlow-0.9B! Experience the magic of instant text-to-image generation here: https://t.co/V9HC88RK0h
arXiv Paper: https://t.co/ndPHJ6Mlxy
Project Page: https://t.co/Cp6MXvu2Zg
We have multiple postdoc openings at the AI Institute for the Foundations of Machine Learning (IFML). Fellows can work with all IFML groups in
UT Austin, Univ. of Washington and Microsoft Research
https://t.co/k33XqTtX8R
(1/3)
Very fortunate to be an intern at #intuit AI!
This precious experience has greatly helped me to bridge the gap between research and deployment. Our developed model has significantly improved performance over the current solution, and will be deployed into multiple pipelines.