With Adam Smith and Jonathan Ullman, we have compiled a set of profiles of 29 people in the "foundations of responsible computing" community ("mathematical research in computation and society writ large") who are on the faculty job market. Check it out in the next tweet! 1/4 👇

taught my first class at penn today. topic: optimization in PyTorch

the inverted final project in my class this semester is a test of how small we can make the gap between "what to do" vs "how to do" with access to llm tools in a topic students know little about.

Our work has been selected as an Oral at ICLR 25! We find theoretical and empirical explanations for the benefits of progressive distillation. Amazing work led by Abhishek Panigrahi and Bingbin Liu, done in collaboration with Andrej Risteski and Surbhi Goel :)

We have a new blog post on reinforcement learning theory for language model post training! By Akshay Krishnamurthy and Audrey Huang (Audrey Huang)! Link in next post 1/2

📢For the first time ever Conference on Learning Theory (COLT) 2025 will feature 1 day dedicated to contributed workshops, tutorials, and community-building events. Conference is on June 30-July 4 in Lyon, France. Submit a proposal by April 18 (AOE). #MLTheory #COLT2025

Super excited about this new paper on (multi-modal/human-AI) collaboration with a fabulous team! TLDR: We show that by communicating only predictions/actions across multiple rounds of interaction, the team can do as well as if each party shared all the information they had.

I won't be at ICLR this year, but I'm sharing my papers here, including an oral in the main conference and several orals in the workshops. Topics include: distillation, data selection for pre-training, benchmark + provable alg for unlearning, pref learning, and overtraining.

I'm at #ICLR2025 🇸🇬 and will be presenting Conformal Language Model Reasoning with Coherent Factuality (arXiv to come soon) this afternoon (4/24, poster session 2, 3-5:30 pm)! This work is with my amazing collaborators Max Rubin-Toles, Maya Gambhir, Aaron Roth, and Surbhi Goel!

Thrilled that Abhishek Panigrahi is presenting our paper (joint with Bingbin Liu, Sadhika Malladi, Andrej Risteski) on the benefits of progressive distillation as an oral at #ICLR2025. Talk details below ⬇️ Also check out our blog post: unprovenalgos.github.io/progressive-di…

Excited to present our paper on a logic-based perspective of LLM jailbreaks with Avishree Khare at @ICLR_conf this Saturday, April 26! Poster #268 in Hall 3+2B at 15:00 Singapore time 📄 arXiv: arxiv.org/abs/2407.00075 🔗 Blog: debugml.github.io/logicbreaks/ \begin{thread}

Misha Khodak, Tanya Marwah, along with myself, Nicholas Boffi and Jianfeng Lu are organizing a COLT 2025 workshop on the Theory of AI for Scientific Computing, to be held on the first day of the conference (June 30).

Super excited to announce our ICML workshop on highlighting the power (and limitations?) of small-scale in the era of large-scale ML. You can submit just a Jupyter notebook, Jupyter notebook + paper, or a survey/position paper. Do submit your work and help us spread the word!

Excited to announce MOSS, our ICML workshop focused on discoveries at small scale! We believe there's tremendous potential & creativity in research done with limited resources and would love to hear your ideas. The submission (due May 22nd) can literally be a Jupyter notebook! :)

Announcing the first workshop on Foundations of Post-Training (FoPT) at COLT 2025! 📝 Soliciting abstracts/posters exploring theoretical & practical aspects of post-training and RL with language models! │ 🗓️ Deadline: May 19, 2025

Register now (first-come first-served) for the "Math of Trustworthy ML workshop" at #LagoMaggiore, Switzerland, Oct 12-16 this year, with a great speaker lineup and the opportunity to present your work as a poster session or contributed talk. Details @ mmlworkshop.ethz.ch

📣Join us at COLT 2025 in Lyon for a community event! 📅When: Mon, June 30 | 16:00 CET What: Fireside chat w/ Peter Bartlett & Vitaly Feldman (Vitaly 🇺🇦 Feldman) on communicating a research agenda, followed by mentorship roundtable to practice elevator pitches & mingle w/ COLT community!

Omw to Vancouver for our workshop. Come check out the amazing talks, posters, and panel.

Excited to announce the Reliable ML workshop at NeurIPS Conference 2025! 📣 How do we build trustworthy models under distribution shift, adversarial attacks, strategic behavior, and missing data? Fantastic lineup of speakers/panelists: Ahmad Beirami (Ahmad Beirami), Surbhi Goel