Cite

Carson, Jack David, and Amir Reisizadeh. A Statistical Physics of Language Model Reasoning. arXiv:2506.04374, arXiv, 4 June 2025. arXiv.org, https://doi.org/10.48550/arXiv.2506.04374.

Metadata

Title: A Statistical Physics of Language Model Reasoning Authors: Jack David Carson, Amir Reisizadeh Cite key: carson2025Statistical

Links

Abstract

Transformer LMs show emergent reasoning that resists mechanistic understanding. We offer a statistical physics framework for continuous-time chain-of-thought reasoning dynamics. We model sentence-level hidden state trajectories as a stochastic dynamical system on a lower-dimensional manifold. This drift-diffusion system uses latent regime switching to capture diverse reasoning phases, including misaligned states or failures. Empirical trajectories (8 models, 7 benchmarks) show a rank-40 projection (balancing variance capture and feasibility) explains ~50% variance. We find four latent reasoning regimes. An SLDS model is formulated and validated to capture these features. The framework enables low-cost reasoning simulation, offering tools to study and predict critical transitions like misaligned states or other LM failures.

Notes

From Obsidian

(As notes and annotations from Zotero are one-way synced, this section include a link to another note within Obsidian to host further notes)

A-Statistical-Physics-of-Language-Model-Reasoning

From Zotero

(one-way sync from Zotero)

Annotations

Highlighting colour codes

  • Note: highlights for quicker reading or comments stemmed from reading the paper but might not be too related to the paper
  • External Insight: Insights from other works but was mentioned in the paper
  • Question/Critic: questions or comments on the content of paper
  • Claim: what the paper claims to have found/achieved
  • Finding: new knowledge presented by the paper
  • Important: anything interesting enough (findings, insights, ideas, etc.) that’s worth remembering
Link to original

From Zotero

(one-way sync from Zotero)