🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

If you're interested in how to keep challenging neural networks throughout training, check out our latest preprint! #sample_efficiency #scaling_laws

Can only speak for my ICML reviewing batch, but the hack of putting scary, convoluted and wrong math still works.

GDM Mech Interp Update: We study if SAEs help probes generalise OOD (they don't 😢). Based on this + parallel negative results on real-world tasks, we're de-prioritising SAE work. Our guess is that SAEs aren't useless, but also aren't a game-changer More + new research in 🧵

FPI workshop off to a great start with Emtiyaz Khan talking about Adaptive Bayesian Intelligence! Come check it out in Peridot 202-203 #FPIWorkshop #ICLR25

Come check out the workshop and hear about novel works and contributions from an exciting lineup of speakers and panelists!

Great talk by Grant Rotskoff linking sampling with nonequilibrium physics in the #FPIworkshop. Come by for the poster session (Peridot 202-203) for the next hour. #ICLR25

that’s a wrap for #AI4Mat and #FPIWorkshop at #ICLR2025 🎁 thanks for coming, off to sleep for the next 2 weeks 😴 #FPIWorkshop: Tara Akhound Sadegh Alex Tong Joey Bose Sarthak Mittal Kirill Neklyudov Yuanqi Du #AI4Mat: Santiago Miret Rocío M N M Anoop Krishnan Stefano Martiniani Mohamad Moosavi

Happy to share that Compositional Risk Minimization has been accepted at #ICML2025 📌Extensive theoretical analysis along with a practical approach for extrapolating classifiers to novel compositions! 📜 arxiv.org/abs/2410.06303

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.

Physics says it's fine to be lazy New preprint on minimum-excess-work guidance: arxiv.org/abs/2505.13375 Check out the thread below 👇

⛵️ Excited to share 𝚂𝙰𝙸𝙻𝙾𝚁: a method for *learning to search* with learned world + reward models to plan in the latent space at test-time. Unlike behavior cloning, 𝚂𝙰𝙸𝙻𝙾𝚁 recovers from mistakes without any additional data, DAgger corrections, or ground truth rewards.

🎉Personal update: I'm thrilled to announce that I'm joining Imperial College London Imperial College London as an Assistant Professor of Computing Imperial Computing starting January 2026. My future lab and I will continue to work on building better Generative Models 🤖, the hardest

Explicit latents or implicit marginalization? at #ICML2025 📌 Tue, 11 am 📍East Exhibition Hall A-B (E-1603) Come check out surprising results on whether explicitly incentivizing learning of correct latents improves generalization over implicitly marginalizing it!