Here is a one-minute summary of Sophie Xhonneux (Sophie Xhonneux)'s Efficient Adversarial Training in LLMs with Continuous Attacks. Come see the spotlight poster at NeurIPS Conference today: Poster Session 3 East, #4702

Presenting two works at NeurIPS Conference workshops today, led by my amazing collaborators. Drop by if you’re curious about issues with current preference datasets, and use of model merging methods for alignment! @MIla_Quebec. Details below 🧵👇

New Paper! 📄 Once a model memorizes an example, it stops learning from it! Our latest work explores this phenomenon and the nuanced interplay between memorization and generalization. Let’s dive in! 🚀🧵

In the third contributed talk of the day, Prakhar Ganesh presented his paper comparing g bias mitigation algorithms in ML!

😡 Absolutely disappointed with Overleaf. My account was deleted without my knowledge, and they’ve done nothing to help me recover it or transfer to my secondary email. Years of work, including all my CVs, SOPs, papers, etc., gone! This is unacceptable. #Overleaf

Undergrads interested in agents, simulations - please do reach out. join our slack: lpm.media.mit.edu/join checkout the research: lpm.media.mit.edu/platform

Internship in our group at Mila - Institut québécois d'IA in reinforcement learning + graphs for reducing energy use in buildings. More info and submit an application by Jan 13 here: https://forms. gle/TCChXnvSAHqzPeMz5 (remove the space - added since Twitter now downweights posts with links)

📢New preprint and data alert! 1/ EPISeg: Automated segmentation of the spinal cord on echo planar images using open-access multi-center data Preprint: doi.org/10.1101/2025.0… Dataset: openneuro.org/datasets/ds005… More in the thread 🧵

How can we make recommender systems more transparent and controllable without sacrificing quality? Introducing TEARS a scrutable RS that replaces numerical user profiles with editable text summaries (accepted The Web Conference). arxiv.org/abs/2410.19302 1/🧵

Looking for an undergrad volunteer who's interested in SSMs + transformers for neural decoding/BCIs at Mila! Strong coding + Pytorch skills are a must. Please DM/email me your CV + interests (priority given to those based in Montréal). Thanks! 🧠🤖

COLM is looking for senior PhDs to serve as program chair assistants. This requires faculty/mentor nomination. If interested, please ask your advisor to nominate you? Have a student that you think is a great fit? Please nominate them. Thanks! forms.gle/GeEpGskcwZi8km…

We’re looking for reviewers for our responsible generative AI workshop at CVPR 2025! If you're interested in contributing, sign up here:

Presenting ✨ 𝐂𝐇𝐀𝐒𝐄: 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐧𝐠 𝐜𝐡𝐚𝐥𝐥𝐞𝐧𝐠𝐢𝐧𝐠 𝐬𝐲𝐧𝐭𝐡𝐞𝐭𝐢𝐜 𝐝𝐚𝐭𝐚 𝐟𝐨𝐫 𝐞𝐯𝐚𝐥𝐮𝐚𝐭𝐢𝐨𝐧 ✨ Work w/ fantastic advisors 🇺🇦 Dzmitry Bahdanau and Siva Reddy Thread 🧵:

Llamas browsing the web look cute, but they are capable of causing a lot of harm! Check out our new Web Agents ∩ Safety benchmark: SafeArena! Paper: arxiv.org/abs/2503.04957

📣Thrilled to announce I’ll join Carnegie Mellon University (CMU Engineering & Public Policy & Language Technologies Institute | @CarnegieMellon) as an Assistant Professor starting Fall 2026! Until then, I’ll be a Research Scientist at AI at Meta FAIR in SF, working with Kamalika Chaudhuri’s amazing team on privacy, security, and reasoning in LLMs!

A great collab with former labmates Anthony GX-Chen & Dongyan! Interesting cognitive limitation in LMs: strong disjunctive bias leads to poor performance on conjunctive causal inference tasks. Mirrors adult human biases—possibly a byproduct of training data priors.

Exciting news: the Fairness, Accountability, Transparency and Ethics (FATE) group at Microsoft Research NYC is hiring a predoctoral fellow!!! 🎉 microsoft.com/en-us/research…

My lab’s contributions at #CVPR2025: -- Organizing VLMs4All - CVPR 2025 Workshop workshop (with 2 challenges) sites.google.com/corp/view/vlms… -- 2 main conference papers (1 highlight, 1 poster) cvpr.thecvf.com/virtual/2025/p… (highlight) cvpr.thecvf.com/virtual/2025/p… (poster) -- 4 workshop papers (2 spotlight talks, 2

📣 Call for a postdoc! We are looking for a postdoctoral researcher to work at the intersection of machine learning and materials science. Find all details here: assets-v2.circle.so/4htxv8ljrkzrxv…

(1/n)🚨You can train a model solving DFT for any geometry almost without training data!🚨 Introducing Self-Refining Training for Amortized Density Functional Theory — a variational framework for learning a DFT solver that predicts the ground-state solutions for different