On the Role of Unobserved Sequences on Sample-based Uncertainty Quantification for LLMs

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Kunitomo-Jacquin, Lucie, Marrese-Taylor, Edison, Fukuda, Ken
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909825937440768
author Kunitomo-Jacquin, Lucie
Marrese-Taylor, Edison
Fukuda, Ken
author_facet Kunitomo-Jacquin, Lucie
Marrese-Taylor, Edison
Fukuda, Ken
contents Quantifying uncertainty in large language models (LLMs) is important for safety-critical applications because it helps spot incorrect answers, known as hallucinations. One major trend of uncertainty quantification methods is based on estimating the entropy of the distribution of the LLM's potential output sequences. This estimation is based on a set of output sequences and associated probabilities obtained by querying the LLM several times. In this paper, we advocate and experimentally show that the probability of unobserved sequences plays a crucial role, and we recommend future research to integrate it to enhance such LLM uncertainty quantification methods.
format Preprint
id arxiv_https___arxiv_org_abs_2510_04439
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle On the Role of Unobserved Sequences on Sample-based Uncertainty Quantification for LLMs
Kunitomo-Jacquin, Lucie
Marrese-Taylor, Edison
Fukuda, Ken
Computation and Language
Quantifying uncertainty in large language models (LLMs) is important for safety-critical applications because it helps spot incorrect answers, known as hallucinations. One major trend of uncertainty quantification methods is based on estimating the entropy of the distribution of the LLM's potential output sequences. This estimation is based on a set of output sequences and associated probabilities obtained by querying the LLM several times. In this paper, we advocate and experimentally show that the probability of unobserved sequences plays a crucial role, and we recommend future research to integrate it to enhance such LLM uncertainty quantification methods.
title On the Role of Unobserved Sequences on Sample-based Uncertainty Quantification for LLMs
topic Computation and Language
url https://arxiv.org/abs/2510.04439