My first tweet

Overcoming temptation: Incentive design for intertemporal choice arxiv.org/abs/2203.05782 We use AI models to help individuals adhere to long-term goals (e.g., retirement savings, weight loss) and avoid giving in to temptation.

Excited to share our work on self-supervised video object representation learning: We introduce SAVi++, a slot-based video model that — for the first time — scales to Waymo Open driving scenes w/o direct supervision. 🖥️ slot-attention-video.github.io/savi++ 📜 arxiv.org/abs/2206.07764 1/7

We are excited to make the jump to complex real-world data with this class of models — and about the potential that slot-based models have for reducing the need for detailed human supervision when learning about the physical world. 6/7

1/ Today we are excited to introduce Phenaki: phenaki.github.io, short-link-to-paper, a model for generating videos from text, with prompts that can change over time, and that is able to generate videos that can be as long as multiple minutes!

Two important breakthroughs from Google AI this week - Imagen Video, a new text-conditioned video diffusion model that generates 1280x768 24fps HD video. And Phenaki, a model which generates long coherent videos for a sequence of text prompts. imagen.research.google/video/

Nature Comms paper: Subtle adversarial image manipulations influence both human and machine perception! We show that adversarial attacks against computer vision models also transfer (weakly) to humans, even when the attack magnitude is small. nature.com/articles/s4146…

🔍 New LLM Research 🔍 Conventional wisdom says that deep neural networks suffer from catastrophic forgetting as we train them on a sequence of data points with distribution shifts. But conventions are meant to be challenged! In our recent paper led by Yanlai Yang, we discovered

The ability to properly contextualize is a core competency of LLMs, yet even the best models sometimes struggle. In a new preprint, we use #MechanisticInterpretability techniques to propose an explanation for contextualization errors: the LLM Race Conditions Hypothesis. [1/9]

Excited to present "Recurrent Complex-Weighted Autoencoders for Unsupervised Object Discovery" at #NeurIPS2024! TL;DR: Our model, SynCx, greatly simplifies the inductive biases and training procedures of current state-of-the-art synchrony models. Thread 👇 1/x.

We are announcing the launch of Airial Travel’s open-to-all beta version for desktop today. Airial is your personal travel agent with AI superpowers which makes planning and booking trips as easy as dreaming them up. airial.travel Me and Sanjeev co-founded Airial

[📜1/9] Does machine unlearning truly erase data influence? Our new paper reveals a critical insight: 'forgotten' information often isn't gone—it's merely dormant, and easily recovered by fine-tuning on just the retain set.

Adrian Weller David Krueger Gintare Karolina Dziugaite Michael C. Mozer Eleni Triantafillou [📜9/9] Check out our paper for more details. Paper: arxiv.org/abs/2505.22310 Code: github.com/shoaibahmed/vi…

🌟To appear in the MechInterp Workshop @ #NeurIPS2025 🌟 Paper: arxiv.org/abs/2509.04466 How do language models (LMs) form representation of new tasks, during in-context learning? We study different types of task representations, and find that they evolve in distinct ways. 🧵1/7

Happy to announce that our work has been accepted to workshops on Multi-turn Interactions and Embodied World Models at #NeurIPS2025! Frontier foundation models are incredible, but how well can they explore in interactive environments? Paper👇 arxiv.org/abs/2412.06438 🧵1/13