AgentReview: Exploring Peer Review Dynamics with LLM Agents

TL;DR

AgentReview is a novel LLM-based framework that simulates peer review processes, revealing biases and latent factors affecting paper decisions, while addressing privacy concerns.

Contribution

This work introduces the first LLM-based peer review simulation framework that disentangles latent factors and mitigates privacy issues.

Findings

37.1% variation in paper decisions due to reviewer biases

Insights supported by sociological theories such as social influence and authority bias

Code available at https://github.com/Ahren09/AgentReview

Abstract

Peer review is fundamental to the integrity and advancement of scientific publication. Traditional methods of peer review analyses often rely on exploration and statistics of existing peer review data, which do not adequately address the multivariate nature of the process, account for the latent variables, and are further constrained by privacy concerns due to the sensitive nature of the data. We introduce AgentReview, the first large language model (LLM) based peer review simulation framework, which effectively disentangles the impacts of multiple latent factors and addresses the privacy issue. Our study reveals significant insights, including a notable 37.1% variation in paper decisions due to reviewers' biases, supported by sociological theories such as the social influence theory, altruism fatigue, and authority bias. We believe that this study could offer valuable insights to improve the design of peer review mechanisms. Our code is available at https://github.com/Ahren09/AgentReview.