Justice in Judgment: Unveiling (Hidden) Bias in LLM-assisted Peer Reviews

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Vasu, Sai Suresh Macharla, Sheth, Ivaxi, Wang, Hui-Po, Binkyte, Ruta, Fritz, Mario
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908996345004032
author Vasu, Sai Suresh Macharla
Sheth, Ivaxi
Wang, Hui-Po
Binkyte, Ruta
Fritz, Mario
author_facet Vasu, Sai Suresh Macharla
Sheth, Ivaxi
Wang, Hui-Po
Binkyte, Ruta
Fritz, Mario
contents The adoption of large language models (LLMs) is transforming the peer review process, from assisting reviewers in writing detailed evaluations to generating entire reviews automatically. While these capabilities offer new opportunities, they also raise concerns about fairness and reliability. In this paper, we investigate bias in LLM-generated peer reviews through controlled interventions on author metadata, including affiliation, gender, seniority, and publication history. Our analysis consistently shows a strong affiliation bias favoring authors from highly ranked institutions. We also identify directional preferences associated with seniority and prior publication record, which can influence acceptance decisions for borderline papers. Gender effects are smaller but present in several models. Notably, implicit biases become more pronounced when examining token-level soft ratings, suggesting that alignment may mask but not fully eliminate underlying preferences
format Preprint
id arxiv_https___arxiv_org_abs_2509_13400
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Justice in Judgment: Unveiling (Hidden) Bias in LLM-assisted Peer Reviews
Vasu, Sai Suresh Macharla
Sheth, Ivaxi
Wang, Hui-Po
Binkyte, Ruta
Fritz, Mario
Computers and Society
Artificial Intelligence
The adoption of large language models (LLMs) is transforming the peer review process, from assisting reviewers in writing detailed evaluations to generating entire reviews automatically. While these capabilities offer new opportunities, they also raise concerns about fairness and reliability. In this paper, we investigate bias in LLM-generated peer reviews through controlled interventions on author metadata, including affiliation, gender, seniority, and publication history. Our analysis consistently shows a strong affiliation bias favoring authors from highly ranked institutions. We also identify directional preferences associated with seniority and prior publication record, which can influence acceptance decisions for borderline papers. Gender effects are smaller but present in several models. Notably, implicit biases become more pronounced when examining token-level soft ratings, suggesting that alignment may mask but not fully eliminate underlying preferences
title Justice in Judgment: Unveiling (Hidden) Bias in LLM-assisted Peer Reviews
topic Computers and Society
Artificial Intelligence
url https://arxiv.org/abs/2509.13400