Saved in:
Bibliographic Details
Main Authors: Lee, Dongryeol, Hwang, Yerin, Kim, Yongil, Park, Joonsuk, Jung, Kyomin
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.20774
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912355616555008
author Lee, Dongryeol
Hwang, Yerin
Kim, Yongil
Park, Joonsuk
Jung, Kyomin
author_facet Lee, Dongryeol
Hwang, Yerin
Kim, Yongil
Park, Joonsuk
Jung, Kyomin
contents In line with the principle of honesty, there has been a growing effort to train large language models (LLMs) to generate outputs containing epistemic markers. However, evaluation in the presence of epistemic markers has been largely overlooked, raising a critical question: Could the use of epistemic markers in LLM-generated outputs lead to unintended negative consequences? To address this, we present EMBER, a benchmark designed to assess the robustness of LLM-judges to epistemic markers in both single and pairwise evaluation settings. Our findings, based on evaluations using EMBER, reveal that all tested LLM-judges, including GPT-4o, show a notable lack of robustness in the presence of epistemic markers. Specifically, we observe a negative bias toward epistemic markers, with a stronger bias against markers expressing uncertainty. This suggests that LLM-judges are influenced by the presence of these markers and do not focus solely on the correctness of the content.
format Preprint
id arxiv_https___arxiv_org_abs_2410_20774
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Are LLM-Judges Robust to Expressions of Uncertainty? Investigating the effect of Epistemic Markers on LLM-based Evaluation
Lee, Dongryeol
Hwang, Yerin
Kim, Yongil
Park, Joonsuk
Jung, Kyomin
Computation and Language
In line with the principle of honesty, there has been a growing effort to train large language models (LLMs) to generate outputs containing epistemic markers. However, evaluation in the presence of epistemic markers has been largely overlooked, raising a critical question: Could the use of epistemic markers in LLM-generated outputs lead to unintended negative consequences? To address this, we present EMBER, a benchmark designed to assess the robustness of LLM-judges to epistemic markers in both single and pairwise evaluation settings. Our findings, based on evaluations using EMBER, reveal that all tested LLM-judges, including GPT-4o, show a notable lack of robustness in the presence of epistemic markers. Specifically, we observe a negative bias toward epistemic markers, with a stronger bias against markers expressing uncertainty. This suggests that LLM-judges are influenced by the presence of these markers and do not focus solely on the correctness of the content.
title Are LLM-Judges Robust to Expressions of Uncertainty? Investigating the effect of Epistemic Markers on LLM-based Evaluation
topic Computation and Language
url https://arxiv.org/abs/2410.20774