Saved in:
Bibliographic Details
Main Authors: Chen, Yi-Pei, Chu, KuanChao, Nakayama, Hideki
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.02863
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911905084342272
author Chen, Yi-Pei
Chu, KuanChao
Nakayama, Hideki
author_facet Chen, Yi-Pei
Chu, KuanChao
Nakayama, Hideki
contents This research investigates the effect of prompt design on dialogue evaluation using large language models (LLMs). While LLMs are increasingly used for scoring various inputs, creating effective prompts for dialogue evaluation remains challenging due to model sensitivity and subjectivity in dialogue assessments. Our study experimented with different prompt structures, altering the sequence of output instructions and including explanatory reasons. We found that the order of presenting reasons and scores significantly influences LLMs' scoring, with a "reason-first" approach yielding more comprehensive evaluations. This insight is crucial for enhancing the accuracy and consistency of LLM-based evaluations.
format Preprint
id arxiv_https___arxiv_org_abs_2406_02863
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LLM as a Scorer: The Impact of Output Order on Dialogue Evaluation
Chen, Yi-Pei
Chu, KuanChao
Nakayama, Hideki
Computation and Language
This research investigates the effect of prompt design on dialogue evaluation using large language models (LLMs). While LLMs are increasingly used for scoring various inputs, creating effective prompts for dialogue evaluation remains challenging due to model sensitivity and subjectivity in dialogue assessments. Our study experimented with different prompt structures, altering the sequence of output instructions and including explanatory reasons. We found that the order of presenting reasons and scores significantly influences LLMs' scoring, with a "reason-first" approach yielding more comprehensive evaluations. This insight is crucial for enhancing the accuracy and consistency of LLM-based evaluations.
title LLM as a Scorer: The Impact of Output Order on Dialogue Evaluation
topic Computation and Language
url https://arxiv.org/abs/2406.02863