cd /news/large-language-models/thermodynamic-signatures-of-reasonin… · home topics large-language-models article
[ARTICLE · art-33565] src=arxiv.org ↗ pub= topic=large-language-models verified=true sentiment=↑ positive

Thermodynamic Signatures of Reasoning: Free-Energy and Spectral-Form-Factor Diagnostics for Hallucination Detection in Large Language Models

Researchers introduced Free-Energy Signatures (Fes), a spectral descriptor for hallucination detection in large language models that treats attention Laplacians as Hamiltonians to extract thermodynamic potentials. The method achieved a +6.5 AUROC improvement over LapEig and +2.4 over GoR-4 across six models and benchmarks, with a training-free detector. A complementary random-matrix-theory analysis revealed that correct generations exhibit Wigner-Dyson spectral statistics while hallucinations show Poisson-like statistics.

read1 min views1 publishedJun 19, 2026

arXiv:2606.19404v1 Announce Type: new Abstract: Hallucination detection in large language models (LLMs) is deployment-critical, and recent work shows that the spectrum of attention-derived graph Laplacians carries strong signal about reasoning quality. Prior spectral diagnostics, however, summarize the Laplacian spectrum by a handful of eigenvalues or hand-picked scalars, leaving most of its structure unused. We propose Free-Energy Signatures (Fes), a spectral descriptor that treats each layer's attention Laplacian as a Hamiltonian and extracts its thermodynamic potentials partition function, free energy, spectral entropy, heat capacity together with the random-matrix-theory (RMT) spectral form factor. We prove three results: (i)~Lipschitz stability of Fes under attention perturbation; (ii)~an expressiveness result showing that Fes enriches finite spectral summaries and approximates moment-derived spectral functionals under explicit regularity and grid-resolution assumptions; and (iii)~a finite-sample PAC bound on the AUROC of a training-free detector built from Fes. Empirically, across six open-weight LLMs and six benchmarks, a lightweight probe on Fes descriptors achieves the strongest aggregate AUROC among attention-spectral baselines, improving over LapEig by $+6.5$ AUROC points and over GoR-4 by $+2.4$ points on average, while requiring no update to the underlying LLM. In the fully unsupervised setting, an RMT-deviation score achieves mean AUROC $0.71$, providing a label-free but weaker detector. A complementary RMT analysis shows that correct generations exhibit more Wigner-Dyson like spectral statistics, whereas hallucinations exhibit more Poisson-like statistics. The anonymized code and config are provided in the supplementary material.

── more in #large-language-models 4 stories · sorted by recency
── more on @free-energy signatures 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/thermodynamic-signat…] indexed:0 read:1min 2026-06-19 ·