If you're at ICLR this week, make sure to check out our work on causal representation learning, which got a spotlight! :)

New work from the great Alexia et al. from the Samsung Lab in Montreal on molecule generation, check it out!

Honored to be selected as a 2024 #STATWunderkind! Grateful for all my mentors & collaborators from Genentech and Stanford Medicine. Learn more about our work here: statnews.com/meet-the-2024-…

This was a really fun collaboration :) Working on this project allowed me to gain a better understanding of the linear representation hypothesis in LLM by formalizing it using ideas from identifiable/causal representation learning. Check it out!

1\ Hi, can I get an unsupervised sparse autoencoder for steering, please? I only have unlabeled data varying across multiple unknown concepts. Oh, and make sure it learns the same features each time! Yes! A freshly brewed Sparse Shift Autoencoder (SSAE) coming right up. 🧶

I'm at #ICLR2025 presenting our work on compositional generalization! (Sat. 10 AM; Hall 3 + Hall 2B, #310) We provide a general and unifying theory of compositional generalization, based on a new principle called interaction asymmetry! 📜 arxiv.org/abs/2411.07784 (See 🧵)

This was a really fun collaboration with folks from the Max Planck Institute in Tuebingen. Make sure to pass by our poster if you are at ICLR!

In this one, we explore the linear properties of next-token predictors like LLMs through the lens of identifiability. I learned quite a bit while working on this project! In collaboration with folks from University of Trento and Copenhagen University!

If you're at ICML on Saturday, check out our workshop paper on identifiable steering from multi-concept shifts in language models!

Finally published: “Top-down and bottom-up neuroscience: overcoming the clash of research cultures” nature.com/articles/s41... Looking for ways to better understand different neuroscientific perspectives and enable productive collaborations