Metraipolis › archive › dc_fe62835b1146
Verbalized and Internal Probabilities Are Coupled in Large Language Models — arXiv:2610.00827
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.00827
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T12:11:17.119Z
- SHA-256 of the archived text
bcb57105ecddb5395650da3d80a9c19578e290bd1e7e2b69240196732087cd7c- Stable URL of this page
https://metraipolis.com/documento/dc_fe62835b1146
How to cite
Verbalized and Internal Probabilities Are Coupled in Large Language Models — arXiv:2610.00827 — arXiv (en), https://arxiv.org/abs/2610.00827 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fe62835b1146 (Metraipolis archive dc_fe62835b1146, sha256 bcb57105ecddb5395650da3d80a9c19578e290bd1e7e2b69240196732087cd7c)
BibTeX
@misc{dcfe62835b1146,
title = {Verbalized and Internal Probabilities Are Coupled in Large Language Models — arXiv:2610.00827},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.00827},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fe62835b1146; sha256 bcb57105ecddb5395650da3d80a9c19578e290bd1e7e2b69240196732087cd7c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00827. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models carry an internal notion of uncertainty in their sampling distribution, i.e., the probabilities they place on generating one answer rather than another. They can also be asked to state a confidence, in words or as a number: a verbalized uncertainty. Prior work suggests that internal probabilities track relative frequencies in the training data, and that verbalized probabilities track explicit probabilistic assertions in the training data. However, we do not know whether these two readouts are aligned, except when frequencies and probabilistic assertions in the training data happen to align. This limits our understanding of when we can use verbalized uncertainties as a proxy for either training data frequencies, or a model's internal distribution. We resolve this gap by systematically exploring how LLMs probability readouts are impacted by training and in-context data, via intervening on the underlying uncertainty sources in the data. We find that both internal and verbalized probability readouts are impacted by both distributional and asserted uncertainty in the training data. Further, we find that verbalized and internal probabilities are aligned beyond what would be expected by independently tracking the same uncertainty sources, suggesting that verbalized probabilities can be used to probe a model's internal distribution. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.00827 — 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 2026-10-06T10:21:28.114Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).