Are we still able to recognize pearls? Machine-driven peer review and the risk to creativity: An explainable RAG-XAI detection framework with markers extraction

TL;DR

This paper presents an explainable detection framework (RAG-XAI) that accurately identifies machine-generated peer reviews to preserve transparency and creativity in scientific evaluation.

Contribution

It introduces a novel, highly effective framework for detecting automated reviews, addressing risks of epistemic homogenization and loss of human judgment in peer review.

Findings

Detection accuracy of 99.61% with XGBoost, Random Forest, LightGBM.

Feature analysis highlights absence of personal signals and repetition as key indicators.

RAG component achieves 90.5% top-1 retrieval accuracy.

Abstract

The integration of large language models (LLMs) into peer review raises a concern beyond authorship and detection: the potential cascading automation of the entire editorial process. As reviews become partially or fully machine-generated, it becomes plausible that editorial decisions may also be delegated to algorithmic systems, leading to a fully automated evaluation pipeline. They risk reshaping the criteria by which scientific work is assessed. This paper argues that machine-driven assessment may systematically favor standardized, pattern-conforming research while penalizing unconventional and paradigm-shifting ideas that require contextual human judgment. We consider that this shift could lead to epistemic homogenization, where researchers are implicitly incentivized to optimize their work for algorithmic approval rather than genuine discovery. To address this risk, we introduce an explainable framework (RAG-XAI) for assessing review quality and detecting automated patterns using markers LLM extractor, aiming to preserve transparency, accountability and creativity in science. The proposed framework achieves near-perfect detection performance, with XGBoost, Random Forest and LightGBM reaching 99.61% accuracy, AUC-ROC above 0.999 and F1-scores of 0.9925 on the test set, while maintaining extremely low false positive rates (<0.23%) and false negative rates (~0.8%). In contrast, the logistic regression baseline performs substantially worse (89.97% accuracy, F1-score 0.8314). Feature importance and SHAP analyses identify absence of personal signals and repetition patterns as the dominant predictors. Additionally, the RAG component achieves 90.5% top-1 retrieval accuracy, with strong same-class clustering in the embedding space, further supporting the reliability of the framework's outputs.