When Your Reviewer is an LLM: Biases, Divergence, and Prompt Injection Risks in Peer Review
Preprint
Abstract
Peer review is the cornerstone of academic publishing, yet the process is increasingly strained by rising submission volumes, reviewer overload, and expertise mismatches. Large language models (LLMs) are now being used as "reviewer aids," raising concerns about their fairness, consistency, and robustness against indirect prompt injection attacks. This paper presents a systematic evaluation of LLMs as academic reviewers. Using a curated dataset of 1,441 papers from ICLR 2023 and NeurIPS 2022, we evaluate GPT-5-mini against human reviewers across ratings, strengths, and weaknesses. The evaluation employs structured prompting with reference paper calibration, topic modeling, and similarity analysis to compare review content. We further embed covert instructions into PDF submissions to assess LLMs' susceptibility to prompt injection. Our findings show that LLMs consistently inflate ratings for weaker papers while aligning more closely with human judgments on stronger contributions. Moreover, while overarching malicious prompts induce only minor shifts in topical focus, explicitly field-specific instructions successfully manipulate specific aspects of LLM-generated reviews. This study underscores both the promises and perils of integrating LLMs into peer review and points to the importance of designing safeguards that ensure integrity and trust in future review processes.Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- The Implications of Insecure Use of Fonts Against PDF Documents and Web PagesIEEE Transactions on Information Forensics and Security, 2025
- AI is transforming peer review — and many scientists are worriedNature, 2025
- When combinations of humans and AI are useful: A systematic review and meta-analysisNature Human Behaviour, 2024
- Can Large Language Models Provide Useful Feedback on Research Papers? A Large-Scale Empirical AnalysisNEJM AI, 2024
- End-User Audits: A System Empowering Communities to Lead Large-Scale Investigations of Harmful Algorithmic BehaviorProceedings of the ACM on Human-Computer Interaction, 2022
- Resource use during systematic review production varies widely: a scoping reviewJournal of Clinical Epidemiology, 2021
- Everyday Algorithm Auditing: Understanding the Power of Everyday Users in Surfacing Harmful Algorithmic BehaviorsProceedings of the ACM on Human-Computer Interaction, 2021
- Supporting robust, rigorous, and reliable reviewing as the cornerstone of our profession: Introducing a competency framework for peer reviewIndustrial and Organizational Psychology, 2020
- The benefits and challenges of using systematic reviews in international development researchJournal of Development Effectiveness, 2012
- Impartial Judgment by the “Gatekeepers” of Science: Fallibility and Accountability in the Peer Review ProcessAdvances in Health Sciences Education, 2003