Saved in:
Bibliographic Details
Main Author: Keuper, Janis
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.10248
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912604012675072
author Keuper, Janis
author_facet Keuper, Janis
contents The ongoing intense discussion on rising LLM usage in the scientific peer-review process has recently been mingled by reports of authors using hidden prompt injections to manipulate review scores. Since the existence of such "attacks" - although seen by some commentators as "self-defense" - would have a great impact on the further debate, this paper investigates the practicability and technical success of the described manipulations. Our systematic evaluation uses 1k reviews of 2024 ICLR papers generated by a wide range of LLMs shows two distinct results: I) very simple prompt injections are indeed highly effective, reaching up to 100% acceptance scores. II) LLM reviews are generally biased toward acceptance (>95% in many models). Both results have great impact on the ongoing discussions on LLM usage in peer-review.
format Preprint
id arxiv_https___arxiv_org_abs_2509_10248
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Prompt Injection Attacks on LLM Generated Reviews of Scientific Publications
Keuper, Janis
Machine Learning
The ongoing intense discussion on rising LLM usage in the scientific peer-review process has recently been mingled by reports of authors using hidden prompt injections to manipulate review scores. Since the existence of such "attacks" - although seen by some commentators as "self-defense" - would have a great impact on the further debate, this paper investigates the practicability and technical success of the described manipulations. Our systematic evaluation uses 1k reviews of 2024 ICLR papers generated by a wide range of LLMs shows two distinct results: I) very simple prompt injections are indeed highly effective, reaching up to 100% acceptance scores. II) LLM reviews are generally biased toward acceptance (>95% in many models). Both results have great impact on the ongoing discussions on LLM usage in peer-review.
title Prompt Injection Attacks on LLM Generated Reviews of Scientific Publications
topic Machine Learning
url https://arxiv.org/abs/2509.10248