Research map / Large language models / AI and society
Is Your Paper Being Reviewed by an LLM? Benchmarking AI Text Detection in Peer Review
Sungduk Yu, Man Luo, Avinash Madasu, Vasudev Lal, Phillip Howard
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use manuscript-based anchor review similarity instead of best baseline (Binoculars) | Anchor · GPT-4o reviews, target FPR 0.1% | TPR: 17.1 -> 63.5 | Improved |
| Anchor vs Binoculars on Gemini reviews | Anchor · Gemini reviews | TPR: 61.5 vs 59.7 at 0.1% FPR | No clear effect |
| Anchor on Llama reviews vs Binoculars | Anchor · Llama reviews, target FPR 0.1% | TPR: 98.4 (Binoculars) vs 13.6 (Anchor) | Hurt |
| calibrate on ICLR+NeurIPS instead of ICLR only (ablation) | Binoculars and Anchor · Gemini reviews, 0.1% FPR | TPR: Anchor 59.7 -> 57.2; Binoculars 61.5 -> 63.8 | Mixed |
| raise LLM editing level of human reviews from Extensive to Maximum (ablation) | Anchor · Edited ICLR 2021 reviews, 0.1% FPR | fraction flagged as AI: 1.9 -> 60.8 | Improved |
| generate reviews from bullet points of human reviews rather than fully with AI | Anchor · ICLR 2021 | TPR at 0.1% target FPR: 63.5 -> 5.8 | Hurt |
| use archetype-based prompts instead of score-aligned prompt (ablation) | AI detectors · ICLR 2021 GPT-4o reviews | TPR | No clear effect |
| prompt LLMs to mimic human reviews and avoid AI characteristics (ablation) | detectors · 6,000 generated reviews | AUROC: AUROC differences <= 0.01 | No clear effect |
| use agent-based generation pipelines (AI Scientist, AgentReview, DeepReview) (ablation) | Binoculars · ICLR 2022 reviews | AUROC: > 0.99 | No clear effect |
Closest papers by meaning
- Monitoring AI-Modified Content at Scale: A Case Study on the Impact of ChatGPT on AI Conference Peer Reviews (ICML 2024 · similarity 0.91)
- From Replication to Redesign: Exploring Pairwise Comparisons for LLM-Based Peer Review (NeurIPS 2025 · similarity 0.89)
- CycleResearcher: Improving Automated Research via Automated Review (ICLR 2025 · similarity 0.88)
- Agent Reviewers: Domain-specific Multimodal Agents with Shared Memory for Paper Review (ICML 2025 · similarity 0.87)
- Position: The Artificial Intelligence and Machine Learning Community Should Adopt a More Transparent and Regulated Peer Review Process (ICML 2025 · similarity 0.87)
- PiCO: Peer Review in LLMs based on Consistency Optimization (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
