Metraipolis › archive › dc_c985814bb609
Mean-Field Dynamics of Chain-of-Thought Reasoning in Large Language Models — arXiv:2608.05152
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.05152
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-25T21:26:59.072Z
- SHA-256 of the archived text
06f42739d8c4825263347a603ec0c68370a4bb3faa7325eb049c0c9403c7c1a4- Stable URL of this page
https://metraipolis.com/documento/dc_c985814bb609
How to cite
Mean-Field Dynamics of Chain-of-Thought Reasoning in Large Language Models — arXiv:2608.05152 — arXiv (en), https://arxiv.org/abs/2608.05152 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c985814bb609 (Metraipolis archive dc_c985814bb609, sha256 06f42739d8c4825263347a603ec0c68370a4bb3faa7325eb049c0c9403c7c1a4)
BibTeX
@misc{dcc985814bb609,
title = {Mean-Field Dynamics of Chain-of-Thought Reasoning in Large Language Models — arXiv:2608.05152},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.05152},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c985814bb609; sha256 06f42739d8c4825263347a603ec0c68370a4bb3faa7325eb049c0c9403c7c1a4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05152. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models (LLMs) with chain-of-thought reasoning have been widely applied in recent years, and theoretical explanations of their behavior may help deepen our understanding and guide model optimization. In this study, we introduce a framework that seeks statistical regularities and theoretical interpretations in LLM reasoning without simplifying the model architecture or making analogies to existing physical systems. We formulate LLM reasoning as a guided discovery process on a clue graph, and derive a one-dimensional ordinary differential equation for the fraction of discovered clues using the mean-field approximation. Experimentally, clue tokens are identified using the normalized surprisal of a student LLM on the outputs of a teacher LLM, and statistical regularities are obtained by averaging over many reasoning chains of thought. Our experiments show that the resulting statistical regularities are reproducible within the same dataset and can be fitted by the solving the proposed theoretical equation. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-07 in cs.CL). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.05152 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-25T21:24:38.160Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).