François Chollet on Twitter / X

Vehicle: Bridging the Embedding Gap in the Verification of...

Neuro-symbolic programs, i.e. programs containing both machine learning components and traditional symbolic code, are becoming increasingly widespread. Finding a general methodology for verifying...

Cognitive Architectures for Language Agents

Recent efforts have augmented large language models (LLMs) with external resources (e.g., the Internet) or internal control flows (e.g., prompt chaining) for tasks requiring grounding or reasoning, leading to a new class of language agents. While these agents have achieved substantial empirical success, we lack a framework to organize existing agents and plan future developments. In this paper, we draw on the rich history of cognitive science and symbolic artificial intelligence to propose Cognitive Architectures for Language Agents (CoALA). CoALA describes a language agent with modular memory components, a structured action space to interact with internal memory and external environments, and a generalized decision-making process to choose actions. We use CoALA to retrospectively survey and organize a large body of recent work, and prospectively identify actionable directions towards more capable agents. Taken together, CoALA contextualizes today’s language agents within the broader history of AI and outlines a path towards language-based general intelligence.

Systems programming the model

This paper examines the status of the language model object in generative AI, arguing that what we call a ‘model’ is inseparable from the systems deploying it. I first theorize how these objects emerge from systems-level interactions between trained artifacts, prompting mechanisms, and sampling methods, drawing on the philosophy of digital objects as well as software studies to show how models gain their objective character. Such interactions converge on programming, not prompting, language models, and I illustrate how critical code studies can therefore track these dynamics. In an overview of language model programming approaches, I discuss how prompt and program converge, demonstrating how this confluence tends toward the production of new feedback loops wherein models become models of and for themselves. Understanding these feedback loops is essential in view of recent efforts to infrastructuralize AI, in which multiple models cascade into compound systems that abstract toward a unified model of models. Thus the need, I argue, for a systems-level view that can address this new order of abstraction and complexity by identifying where and how the model emerges from the system.

Eliciting Reasoning in Language Models with Cognitive Tools

The recent advent of reasoning models like OpenAI's o1 was met with excited speculation by the AI community about the mechanisms underlying these capabilities in closed models, followed by a rush of replication efforts, particularly from the open source community. These speculations were largely settled by the demonstration from DeepSeek-R1 that chains-of-thought and reinforcement learning (RL) can effectively replicate reasoning on top of base LLMs. However, it remains valuable to explore alternative methods for theoretically eliciting reasoning that could help elucidate the underlying mechanisms, as well as providing additional methods that may offer complementary benefits. Here, we build on the long-standing literature in cognitive psychology and cognitive architectures, which postulates that reasoning arises from the orchestrated, sequential execution of a set of modular, predetermined cognitive operations. Crucially, we implement this key idea within a modern agentic tool-calling framework. In particular, we endow an LLM with a small set of "cognitive tools" encapsulating specific reasoning operations, each executed by the LLM itself. Surprisingly, this simple strategy results in considerable gains in performance on standard mathematical reasoning benchmarks compared to base LLMs, for both closed and open-weight models. For instance, providing our "cognitive tools" to GPT-4.1 increases its pass@1 performance on AIME2024 from 32% to 53%, even surpassing the performance of o1-preview. In addition to its practical implications, this demonstration contributes to the debate regarding the role of post-training methods in eliciting reasoning in LLMs versus the role of inherent capabilities acquired during pre-training, and whether post-training merely uncovers these latent abilities.

Agentic AI gets lost

On the failure of AI to develop world models

JUMPERZ on Twitter / X

karpathy is showing one of the simplest AI architectures that actually works..dump research into a folder, let the model organise it into a wiki, ask questions, then file the answers back in.the real insight is the loop...every query makes the wiki better. it compounds.. now… https://t.co/vRhImgTQuN pic.twitter.com/Uq2D3ONvGT— JUMPERZ (@jumperz) April 2, 2026

SymbolicAI: A Neuro-Symbolic Perspective on Large Language Models (LLMs)

A neurosymbolic perspective on LLMs

</> htmx ~ Working With AI: A Concrete Example

In this essay, Carson Gross walks through a concrete bug fix in hyperscript to show where AI helped, where it fell short, and why keeping a knowledgeable human in the loop is what kept complexity in check.

AI That Evolves in the Wild | Edge.org

I’m interested not in domesticated AI—the stuff that people are trying to sell. I'm interested in wild AI—AI that evolves in the wild. I’m a naturalist, so that’s the interesting thing to me. Thirty-four years ago there was a meeting just like this in which Stanislaw Ulam said to everybody in the room—they’re all mathematicians—"What makes you so sure that mathematical logic corresponds to the way we think?" It’s a higher-level symptom. It’s not how the brain works. All those guys knew fully well that the brain was not fundamentally logical.

An Interpretable Automated Mechanism Design Framework with Large Language Models

Mechanism design has long been a cornerstone of economic theory, with traditional approaches relying on mathematical derivations. Recently, automated approaches, including differentiable economics with neural networks, have emerged for designing payments and allocations. While both analytical and automated methods have advanced the field, they each face significant weaknesses: mathematical derivations are not automated and often struggle to scale to complex problems, while automated and especially neural-network-based approaches suffer from limited interpretability. To address these challenges, we introduce a novel framework that reformulates mechanism design as a code generation task. Using large language models (LLMs), we generate heuristic mechanisms described in code and evolve them to optimize over some evaluation metrics while ensuring key design criteria (e.g., strategy-proofness) through a problem-specific fixing process. This fixing process ensures any mechanism violating the design criteria is adjusted to satisfy them, albeit with some trade-offs in performance metrics. These trade-offs are factored in during the LLM-based evolution process. The code generation capabilities of LLMs enable the discovery of novel and interpretable solutions, bridging the symbolic logic of mechanism design and the generative power of modern AI. Through rigorous experimentation, we demonstrate that LLM-generated mechanisms achieve competitive performance while offering greater interpretability compared to previous approaches. Notably, our framework can rediscover existing manually designed mechanisms and provide insights into neural-network based solutions through Programming-by-Example. These results highlight the potential of LLMs to not only automate but also enhance the transparency and scalability of mechanism design, ensuring safe deployment of the mechanisms in society.

Introducing Claude Fable 5.1 and Claude Mythos 5.1

Our most advanced models for coding and knowledge work. Their research capabilities also offer an early glimpse of how AI models will contribute to scientific progress.

Crafting a good (reasoning) model

A recent talk I gave on model training, reasoning, and the next frontier.

Large AI models are cultural and social technologies – Henry Farrell

Debates about artificial intelligence (AI) tend to revolve around whether large models are intelligent, autonomous agents. Some AI researchers and commentators speculate that we are on the cusp of creating agents with artificial general intelligence (AGI), a prospect anticipated with both elation and anxiety. There have also been extensive conversations about cultural and social consequences of large models, orbiting around two foci: immediate effects of these systems as they are currently used, and hypothetical futures when these systems turn into AGI agents perhaps even superintelligent AGI agents.