ChatGPT: Excellent Paper! Accept It. Editor: Imposter Found! Review Rejected

2025-12-23

AI Summary

Misuse of large language models (LLMs) in scientific writing and peer review is precipitating an academic trust crisis: generated content often lacks originality, fabricates results, and embeds implicit biases—potentially misleading downstream research and compromising safety-critical domains such as healthcare. This paper introduces the first bidirectional LLM adversarial framework tailored to academic review. On the attack side, it employs PDF steganography for implicit prompt injection to manipulate LLM review judgments. On the defense side, it innovatively repurposes prompt injection techniques as detection tools, establishing an “injection–detection” mechanism that identifies spurious reviews via trigger-word analysis. The method integrates steganographic embedding, behavioral pattern analysis, adversarial prompt engineering, and interpretable log auditing. Experiments across mainstream LLMs achieve >92% accuracy in detecting fabricated reviews. We further propose a deployable journal alert protocol, balancing technical feasibility with ethical safeguards.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Large Language Models (LLMs) like ChatGPT are now widely used in writing and reviewing scientific papers. While this trend accelerates publication growth and reduces human workload, it also introduces serious risks. Papers written or reviewed by LLMs may lack real novelty, contain fabricated or biased results, or mislead downstream research that others depend on. Such issues can damage reputations, waste resources, and even endanger lives when flawed studies influence medical or safety-critical systems. This research explores both the offensive and defensive sides of this growing threat. On the attack side, we demonstrate how an author can inject hidden prompts inside a PDF that secretly guide or "jailbreak" LLM reviewers into giving overly positive feedback and biased acceptance. On the defense side, we propose an "inject-and-detect" strategy for editors, where invisible trigger prompts are embedded into papers; if a review repeats or reacts to these triggers, it reveals that the review was generated by an LLM, not a human. This method turns prompt injections from vulnerability into a verification tool. We outline our design, expected model behaviors, and ethical safeguards for deployment. The goal is to expose how fragile today's peer-review process becomes under LLM influence and how editorial awareness can help restore trust in scientific evaluation.

Problem

• Detect LLM-generated reviews in scientific publishing
• Expose hidden prompt injections that bias peer review
• Restore trust in peer review by identifying automated feedback

Innovation

• Inject hidden prompts to jailbreak LLM reviewers
• Embed invisible triggers to detect LLM-generated reviews
• Turn prompt injections into verification tool