Contrastive Decoding Mitigates Score Range Bias in LLM-as-a-Judge

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autor principal: Fujinuma, Yoshinari
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918433855111168
author Fujinuma, Yoshinari
author_facet Fujinuma, Yoshinari
contents Large Language Models (LLMs) are commonly used as evaluators in various applications, but the reliability of the outcomes remains a challenge. One such challenge is using LLMs-as-judges for direct assessment, i.e., assigning scores from a specified range without any references. Focusing on summarization, we first show that this challenge stems from LLM judge outputs being associated with score range bias, i.e., LLM judge outputs are highly sensitive to pre-defined score ranges. We also show that similar biases exist among models from the same family. We then mitigate this bias through contrastive decoding, achieving up to 11.7% relative improvement on average in Spearman correlation with human judgments across different score ranges.
format Preprint
id arxiv_https___arxiv_org_abs_2510_18196
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Contrastive Decoding Mitigates Score Range Bias in LLM-as-a-Judge
Fujinuma, Yoshinari
Computation and Language
Artificial Intelligence
Large Language Models (LLMs) are commonly used as evaluators in various applications, but the reliability of the outcomes remains a challenge. One such challenge is using LLMs-as-judges for direct assessment, i.e., assigning scores from a specified range without any references. Focusing on summarization, we first show that this challenge stems from LLM judge outputs being associated with score range bias, i.e., LLM judge outputs are highly sensitive to pre-defined score ranges. We also show that similar biases exist among models from the same family. We then mitigate this bias through contrastive decoding, achieving up to 11.7% relative improvement on average in Spearman correlation with human judgments across different score ranges.
title Contrastive Decoding Mitigates Score Range Bias in LLM-as-a-Judge
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2510.18196