Researchers have developed a novel system called xPeer, designed to simulate scholarly peer review using AI. This system was evaluated against human reviews on a dataset of 352 simulation records across various disciplines. While xPeer produced longer, more detailed reports with higher targeting and category coverage, human reviews demonstrated stronger explicit reasoning and lexical attestation to manuscripts. The study found low agreement between AI and human reviews in terms of concerns raised and recommendations made, highlighting distinct review profiles and establishing a baseline for reproducibility. AI
IMPACT Establishes a benchmark for AI-driven peer review, potentially impacting academic publishing workflows and evaluation methodologies.
RANK_REASON Academic paper detailing a new benchmark and simulation engine for scholarly peer review. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
