In-context learning emerges in chemical reaction networks without attention

Carlos Floyd; Hector Manuel Lopez Rios; Aaron R. Dinner; Suriyanarayanan Vaikuntanathan

Home/Archive/arXiv

arXiv · 2601.06712

In-context learning emerges in chemical reaction networks without attention

Carlos Floyd·Hector Manuel Lopez Rios·Aaron R. Dinner·Suriyanarayanan Vaikuntanathan

View original source ↗Read source document ↗Export metadata ↓

Abstract

We investigate whether chemical processes can perform in-context learning (ICL), a mode of computation typically associated with transformer architectures. ICL allows a system to infer task-specific rules from a sequence of examples without relying solely on fixed parameters. Traditional ICL relies on a pairwise attention mechanism which is not obviously implementable in chemical systems. However, we show theoretically and numerically that chemical processes can achieve ICL through a mechanism we call subspace projection, in which the entire input vector is mapped onto comparison subspaces, with the dominant projection determining the computational output. We illustrate this mechanism analytically in small chemical systems and show numerically that performance is robust to input encoding and dynamical choices, with the number of tunable degrees of freedom in the input encoding as a key limitation. Our results provide a blueprint for realizing ICL in chemical or other physical media and suggest new directions for designing adaptive synthetic chemical systems and understanding possible biological computation in cells.

Explore related subjects

cond-mat.dis-nn cond-mat.stat-mech q-bio.MN

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Carlos Floyd, Hector Manuel Lopez Rios, Aaron R. Dinner, Suriyanarayanan Vaikuntanathan. 2026-01-10. In-context learning emerges in chemical reaction networks without attention. https://arxiv.org/abs/2601.06712

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

The critical slowing down in training diffusion models

Computational sampling has been central to the sciences since the mid-20th century. While machine-learning-based approaches have recently enabled major advances, their behavior remains poorly understood, with limited theoretical control over when and why they succeed. Here we provide such insight for diffusion models---a class of generative schemes highly effective in practice---by analyzing their application to the $O(n)$ model of statistical field theory in the Gaussian limit $n \to \infty$. In this analytically tractable setting, we show that training a score model with a one-layer network architecture matching the exact solution exhibits a form of critical slowing down in parameter learning. This slowing down also impacts the generation process, indicating that the well-known difficulties of sampling near criticality persist even for learned generative models. To overcome this bottleneck, we consider the power of architectural depth. We find that using a two-layer architecture drastically reduces the critical slowing down, with the training time scaling logarithmically rather than quadratically with system size. Using a Fourier implementation of the architecture, we further show that this acceleration in training time can be achieved without drastically increasing operational complexity. Taken together, these results demonstrate that diffusion models can overcome the critical slowing down through appropriate architectural design, and establish a controlled framework for understanding and improving learned sampling methods in statistical physics and beyond.

cond-mat.dis-nn↗

arXiv2026

Switching diffusivity selects Pareto tail exponent in random growth with redistribution

Random multiplicative growth with redistribution generates stationary Pareto wealth tails in the Bouchaud-Mézard model, but assumes a fixed multiplicative noise intensity. This is restrictive for physical and financial growth processes, where volatility (diffusivity) is often fluctuating. We replace the constant noise intensity by a switching diffusivity and ask how these fluctuations select the Pareto stationary tail. For a geometric Brownian motion with switching diffusivity, the long-time Gaussian limit holds when the redraw law has finite mean and variance. The asymptotic variance retains a contribution from diffusivity persistence. With redistribution and a general redraw law, the stationary large-wealth problem is characterized by a spectral condition for admissible algebraic modes. For a two-state diffusivity, an exact tail analysis gives a Pareto exponent interpolating between the high-diffusivity slow-refresh limit and the mean-diffusivity fast-refresh Bouchaud-Mézard limit.

cond-mat.dis-nn↗

arXiv2026

Signatures of Nonergodicity in Sparse Random Matrices

The prevalence of sparsity in the Fock space graph of interacting many-body systems motivates an investigation into the spectral statistics of sparse random matrices with on-site disorder. We numerically determine the delocalization-localization transition in the ground state as a function of the sparsity. The short-range energy correlation in the bulk indicates that the Anderson transition at infinite temperature occurs at the critical percolation limit of the sparse graph. By analytically deriving the energy moments and calculating the shifted kurtosis, we show that the critical sparsity threshold matches the Anderson transition. Furthermore, long-range energy correlations in the bulk spectrum reveal a Thouless energy scale, suggesting a broad nonergodic regime within the delocalized phase.

cond-mat.dis-nn↗