Leonard Read's pencil story: no one knows how to make a pencil from scratch, yet millions are made without a central manager. Each worker knows only their part. Incentives, not central control, coordinate their efforts. Current multi-agent AI seems to work the opposite way. A central manager does everything: assigns roles, writes the workflow, and decides who acts next. As agents increase and tasks get harder, that manager might become the bottleneck: it should understand every agent and every task, and adaptation becomes inefficient. So how can a group of machine agents organize themselves into something none of them could achieve alone? Further, how can that society of agents keep improving itself over time? In Economy of Minds (EoM), we show that simple local incentives can drive global self-coordination and self-improvement for modern LLM agents. We let agents live in a minimal economy: they compete for work, earn by helping others succeed, and go broke when they don't. Workflows, specialists, and improvement naturally emerge, with no one in charge. Across 5 domains (math reasoning, financial research, scientific research, accelerator design, and distributed-system optimization), limited agents grow into strong societies. The whole idea in a 90-second cartoon👇

Harvard SEAS has a search out for a tenure-track CS prof this year. You can email me if you have any questions.

Today, we introduce Priming Models, a new paradigm for training one-step generative models from scratch. No diffusion teacher. No flow map. No CFG. No JVP. We believe this work will challenge how we think about building one-step models. Joint work with @sitanch, @du_yilun, @parkwbin03, Bohyung Han. 🔥 1.19 FID one-step on ImageNet 256×256 while using substantially less compute, e.g., 12X less than Drifting-B, 6X less than Drifting-L 🎨 Scales to text-to-image generation (GenEval 0.82) 🤖 World modeling on RT-1 / Bridge-V2 Priming Model is built on one core message: [Reconstruction pretraining --> Distributional post-training].

How much representational diversity exists around a trained network? A lot! Moving a ViT’s weights by 1.3% while preserving its predictions, its representations end up less similar to the original than a random-init ViT’s ‼️ New paper: Hessian Null Space Continuation (HNC) 💫🧵 (1/12)

New Preprint! 🚨 Looks can be deceptive! Language models store numbers internally in pretty shapes like curves and helices. Do they compute with these shapes? Not always! To compare two numbers, we discover a neat algorithm: use lines per number and combine them geometrically!

SFT is not dead! 🥳 We found a way to make SFT rival current prevailing posttraining methods, often generalizing better and forgetting less than RL and OPSD. 🤯 Following our prior work on reasoning with sampling, we now introduce sampling to the posttraining stack. 1/n

When do biological reasoning models use their biological inputs? High performance can create a mirage of reliance on biological representations. Reasoning traces don’t always reflect the evidence. Our paper examines these limits and paths to better biological reasoning models.

AI titan Anthropic has launched a biology ‘wet lab’ where human scientists and AI agents will work together to design and conduct experiments. The company has released one of the team’s first finds: a peculiar pattern of DNA in the genomes of several giant viruses. go.nature.com/46K4qmC

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]