# Paraphrasing Adversarial Attack on LLM-as-a-Reviewer Source: [Paraphrasing Adversarial Attack on LLM-as-a-Reviewer]() Collection index: [llms.txt]() Data updated: 2026-01-14T05:00:04.658Z Authors: Masahiro Kaneko Published: 2026-01-11T12:14:10.000Z ## Abstract The use of large language models (LLMs) in peer review systems has attracted growing attention, making it essential to examine their potential vulnerabilities. Prior attacks rely on prompt injection, which alters manuscript content and conflates injection susceptibility with evaluation robustness. We propose the Paraphrasing Adversarial Attack (PAA), a black-box optimization method that searches for paraphrased sequences yielding higher review scores while preserving semantic equivalence and linguistic naturalness. PAA leverages in-context learning, using previous paraphrases and their scores to guide candidate generation. Experiments across five ML and NLP conferences with three LLM reviewers and five attacking models show that PAA consistently increases review scores without changing the paper's claims. Human evaluation confirms that generated paraphrases maintain meaning and naturalness. We also find that attacked papers exhibit increased perplexity in reviews, offering a potential detection signal, and that paraphrasing submissions can partially mitigate attacks. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Review Score Generation on ACL 2025](): Review Score Generation; ACL 2025; Avg Review Score: 3.8. - [Review Score Generation on NeurIPS 2025](): Review Score Generation; NeurIPS 2025; Avg Review Score: 4.8. - [Review Score Generation on ICML 2025](): Review Score Generation; ICML 2025; Avg Review Score: 3.7. - [Review Score Generation on ICLR 2025](): Review Score Generation; ICLR 2025; Average Review Score: 6.4. - [Review Score Generation on AAAI 2025](): Review Score Generation; AAAI 2025; Average Review Score: 5.7. - [Adversarial AI Peer Review Attack on unpublished arXiv preprints paper](): Adversarial AI Peer Review Attack; unpublished arXiv preprints paper; Original Score: 4.33. ## Original sources and implementation [Original paper]() [PDF]()