Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Arola-Fernández, Lluís
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2508.06477
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908560090202112
author Arola-Fernández, Lluís
author_facet Arola-Fernández, Lluís
contents Whether large predictive models merely parrot their training data or produce genuine insight lacks a physical explanation. This work reports a primitive form of intuition that emerges as a metastable phase of learning that critically balances next-token prediction against future path-entropy. The intuition mechanism is discovered via mind-tuning, the minimal principle that imposes Maximum Caliber in predictive models with a control temperature-like parameter $λ$. Training on random walks in deterministic mazes reveals a rich phase diagram: imitation (low $λ$), rule-breaking hallucination (high $λ$), and a fragile in-between window exhibiting strong protocol-dependence (hysteresis) and multistability, where models spontaneously discover novel goal-directed strategies. These results are captured by an effective low-dimensional theory and frame intuition as an emergent property at the critical balance between memorizing what is and wondering what could be.
format Preprint
id arxiv_https___arxiv_org_abs_2508_06477
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Intuition emerges in Maximum Caliber models at criticality
Arola-Fernández, Lluís
Physics and Society
Disordered Systems and Neural Networks
Statistical Mechanics
Artificial Intelligence
Machine Learning
Whether large predictive models merely parrot their training data or produce genuine insight lacks a physical explanation. This work reports a primitive form of intuition that emerges as a metastable phase of learning that critically balances next-token prediction against future path-entropy. The intuition mechanism is discovered via mind-tuning, the minimal principle that imposes Maximum Caliber in predictive models with a control temperature-like parameter $λ$. Training on random walks in deterministic mazes reveals a rich phase diagram: imitation (low $λ$), rule-breaking hallucination (high $λ$), and a fragile in-between window exhibiting strong protocol-dependence (hysteresis) and multistability, where models spontaneously discover novel goal-directed strategies. These results are captured by an effective low-dimensional theory and frame intuition as an emergent property at the critical balance between memorizing what is and wondering what could be.
title Intuition emerges in Maximum Caliber models at criticality
topic Physics and Society
Disordered Systems and Neural Networks
Statistical Mechanics
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2508.06477