New on arXiv
Sort: by field | by arXiv number | trending
Learning Time-Varying Connectomes for Stable Behavior Across Development
Neural circuits must maintain functionality to drive effective behavior while their size, topology, geometry, and intrinsic dynamics change across development. To model developing visuomotor circuits and behavior, we investigated whether a single compact generative encoding could produce functional recurrent controllers across such changes. Building on the HyperNEAT algorithm, we represented this...
Cell-type-specific modularity in E/I networks shapes dynamical regimes and computational capabilities
Structural modularity is a ubiquitous feature of brain networks that shapes neural dynamics and information processing. Yet excitatory and inhibitory populations can differ in their modular organization, and it remains unclear how this asymmetry affects network dynamics and computation. Here, using dynamical mean-field theory and simulations of networks performing complex temporal tasks, we show...
Synchronized Hearts, Desynchronized Brains: A Nonlinear Dynamics Perspective for Function and Failure
Both the heart and the brain are built from electrically excitable cells, yet the emergent dynamics of these organs could not differ more. In the myocardium, a coherent traveling wave of action potentials ensures coordinated contraction; in the cerebral cortex, distributed, irregular spiking underlies efficient information processing and the generation of behavior. Here, we illuminate the...
Mind Control of LLMs via Closed-Loop Stimulation: Out-of-Distribution Robustness
Brains and large language models need not be considered equivalent systems to reveal an interpretable commonality: both exhibit high-dimensional, nonlinear dynamics whose behavior can only be modeled imperfectly. This creates a shared challenge whenever we seek to intervene in their dynamics. A controller may either rely on a model identified around a particular operating regime and risk failure...
A minimal E/I circuit model explains differential recovery from amblyopia under TTX and reverse occlusion
Recovery from amblyopia beyond the critical period remains a major challenge in visual neuroscience. Experimental studies show that transient inactivation of the fellow eye with tetrodotoxin (TTX) produces stronger and more persistent recovery than reverse occlusion (RO), despite both interventions reducing competition from the fellow eye. The circuit mechanisms underlying this difference remain...
Do agents need to sleep?
LLM-based agents, which increasingly operate persistently across sessions and accumulate memory through interaction, already exhibit the familiar symptoms of unchecked plasticity: semantic drift, degraded retrieval precision, propagation of stored errors, and saturation of memory capacity. Drawing on the selectionist "ringing, resonance and renormalization" (RRR) account of sleep, we propose that...
Only titles, authors and the start of each abstract are shown here, as arXiv published them. Open a paper to read its whole abstract and, with your own AI key, a plain-language summary. Papers on arXiv are preprints and may not have been peer reviewed.
Data from arXiv, used through the arXiv API. Thank you to arXiv for use of its open access interoperability.