Saved in:
Bibliographic Details
Main Authors: Wu, Wenqing, Xi, Haixu, Zhang, Chengzhi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.15031
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918028016353280
author Wu, Wenqing
Xi, Haixu
Zhang, Chengzhi
author_facet Wu, Wenqing
Xi, Haixu
Zhang, Chengzhi
contents Peer review is vital in academia for evaluating research quality. Top AI conferences use reviewer confidence scores to ensure review reliability, but existing studies lack fine-grained analysis of text-score consistency, potentially missing key details. This work assesses consistency at word, sentence, and aspect levels using deep learning and NLP conference review data. We employ deep learning to detect hedge sentences and aspects, then analyze report length, hedge word/sentence frequency, aspect mentions, and sentiment to evaluate text-score alignment. Correlation, significance, and regression tests examine confidence scores' impact on paper outcomes. Results show high text-score consistency across all levels, with regression revealing higher confidence scores correlate with paper rejection, validating expert assessments and peer review fairness.
format Preprint
id arxiv_https___arxiv_org_abs_2505_15031
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Are the confidence scores of reviewers consistent with the review content? Evidence from top conference proceedings in AI
Wu, Wenqing
Xi, Haixu
Zhang, Chengzhi
Computation and Language
Artificial Intelligence
Human-Computer Interaction
Information Retrieval
Peer review is vital in academia for evaluating research quality. Top AI conferences use reviewer confidence scores to ensure review reliability, but existing studies lack fine-grained analysis of text-score consistency, potentially missing key details. This work assesses consistency at word, sentence, and aspect levels using deep learning and NLP conference review data. We employ deep learning to detect hedge sentences and aspects, then analyze report length, hedge word/sentence frequency, aspect mentions, and sentiment to evaluate text-score alignment. Correlation, significance, and regression tests examine confidence scores' impact on paper outcomes. Results show high text-score consistency across all levels, with regression revealing higher confidence scores correlate with paper rejection, validating expert assessments and peer review fairness.
title Are the confidence scores of reviewers consistent with the review content? Evidence from top conference proceedings in AI
topic Computation and Language
Artificial Intelligence
Human-Computer Interaction
Information Retrieval
url https://arxiv.org/abs/2505.15031