Imagine a population of machine agents. Each might be strong on certain tasks but fundamentally limited: partial tools, partial observations, finite context, bounded compute. How can these agents self-orchestrate and self-evolve into stronger collective intelligence to solve tasks beyond any single agent's capability? Instead of designing the multi-agent system itself, we propose designing the incentives that govern it. We put agents in an economy. They compete, trade, get wealthy, go bankrupt, and mutate, forming an alive society where coordination and adaptation automatically emerge in a decentralized manner.

1/9 Introducing LOTION (Low-precision optimization via stochastic-noise smoothing), a principled alternative to Quantization-Aware Training (QAT) that explicitly smooths the quantized loss surface while preserving all global minima of the true quantized loss. Details below:

Check out our new work, to be presented at @COLM_conf! We find mechanistic evidence that creativity and hallucinations are tied together. Working on this made me realize that some of the same mechanisms that make LLMs useful may also create failures.

🚨 We finally have a better understanding of how the different paradigms for discrete diffusion compare! In new work with Liye Wang, we prove a quadratic separation between the parallel generation capabilities of uniform and Gaussian diffusion relative to unmasking! 1/

Biology does not need models that just describe the world. It needs models that predict what happens when we intervene. In @CellCellPress Focus on AI in Biology, we lay out how to build biomedical world models, from the data they need to how they should be trained and evaluated.

New episode of #WiAIRpodcast: @GretaTuckute (@KempnerInst @Harvard) talks to @J_Novikova_NLP about what brain-LLM alignment does and does not tell us, and why, asked to keep one label out of neuroscience, cognitive science and AI, she drops AI first

Is your gating mistuned? In our new work, we highlight the critical role of gating for SSMs. Correct gating boosts areas of traditional weakness: in-context learning, retrieval, and long-context performance. arxiv.org/abs/2609.16540 [1/6]

How can we improve robot generalization beyond collecting more demonstrations? A robot can use inference to work out how to solve tasks it was never explicitly trained to perform. With learned models of the world, we can plan our future actions and goals before acting. I wrote a research perspective, Generalization by Construction, on how learning can be combined with inference for flexible generalization. yilundu.github.io/perspectiv…

I will be giving a talk @KempnerInst "Learning Dynamics" workshop today. "When Prediction Shapes the Machine: Evolutionary Reservoirs, Chaos, and Structural Constraints" Enjoyed @SuryaGanguli 's talk and looking forward to see what @dileeplearning @TonyZador and Leslie Valiant have to say kempnerinstitute.harvard.edu…

Very happy to share that I have started as a Kempner Research Fellow at @KempnerInst, Harvard University! Excited to keep exploring diffusion, flows, and new directions in generative modeling, and looking forward to new ideas, collaborations, and getting to know Boston!

At Harvard this week for what promises to be a fun event @KempnerInst !! Learning dynamics in natural and artificial intelligence.

Closed-loop AI scientists AI can generate hypotheses much faster than we can experimentally verify them. @AdaFang_ In our latest piece on Closing the Loop in AI-Driven Biomedical Discovery we outline a path for closed-loop AI scientists, preprints.org/manuscript/202… AI scientists can generate hypotheses, propose experiments, and analyze data, and several have produced findings that laboratories then confirmed. What would it take to close that loop more generally?

Haim Sompolinsky Awarded 2026 Dirac Medal mcb.harvard.edu/department/n… @KempnerInst @Harvard

Code for "Large Language Models Generate Harmful Responses Using a Distinct Mechanism, Shared Across Harm Types" is now available! It includes B-TAP, the interp method behind our work, for localizing and causally intervening on critical weights. Code + project page + paper 👇🏻