Breaking the Reviewer: Assessing the Vulnerability of Large Language Models in Automated Peer Review Under Textual Adversarial Attacks
It addresses the problem of ensuring AI-assisted peer review integrity for academic communities, highlighting risks that could undermine scholarly communication if not mitigated.
This paper investigated the vulnerability of large language models (LLMs) used for automated peer review to textual adversarial attacks, finding that text manipulations can significantly distort LLM assessments and compromise reliability.
Peer review is essential for maintaining academic quality, but the increasing volume of submissions places a significant burden on reviewers. Large language models (LLMs) offer potential assistance in this process, yet their susceptibility to textual adversarial attacks raises reliability concerns. This paper investigates the robustness of LLMs used as automated reviewers in the presence of such attacks. We focus on three key questions: (1) The effectiveness of LLMs in generating reviews compared to human reviewers. (2) The impact of adversarial attacks on the reliability of LLM-generated reviews. (3) Challenges and potential mitigation strategies for LLM-based review. Our evaluation reveals significant vulnerabilities, as text manipulations can distort LLM assessments. We offer a comprehensive evaluation of LLM performance in automated peer reviewing and analyze its robustness against adversarial attacks. Our findings emphasize the importance of addressing adversarial risks to ensure AI strengthens, rather than compromises, the integrity of scholarly communication.