CoCoReviewBench: A Completeness- and Correctness-Oriented Benchmark for AI Reviewers
For researchers developing AI review systems, this benchmark provides a more reliable evaluation method that addresses the unreliability of human reviews as gold standards.
The authors built a benchmark (CoCoReviewBench) for evaluating AI reviewers, focusing on completeness and correctness rather than overlap with human reviews. They curated 3,900 papers from ICLR and NeurIPS and found that AI reviewers are limited in correctness and prone to hallucinations, with reasoning models performing better.
Despite the rapid development of AI reviewers, evaluating such systems remains challenging: metrics favor overlap with human reviews over correctness. However, since human reviews often cover only a subset of salient issues and sometimes contain mistakes, they are unreliable as gold references. To address this, we build category-specific benchmark subsets and skip evaluation when the corresponding human reviews are missing to strengthen Completeness. We also leverage reviewer--author--meta-review discussions as expert annotations and filter unreliable reviews accordingly to strengthen Correctness. Finally, we introduce CoCoReviewBench, which curates 3,900 papers from ICLR and NeurIPS to enable reliable and fine-grained evaluation of AI reviewers. Analysis shows that AI reviewers remain limited in correctness and are prone to hallucinations, and highlights reasoning models as more effective reviewers, motivating further directions for improving AI reviewers. Benchmarks and models are available at https://github.com/hexuandeng/CoCoReviewBench.