PRISM: A Multi-Dimensional Benchmark for Evaluating LLM Peer Reviewers 文章

ArXiv CS.CL2026-05-27NEWSen作者: Ngoc Phan Phuoc Loc, Toan Huynh La Viet, Thanh Tran Khanh, Duy A Nguyen, Tuan Anh Nguyen Pham, Thanh Nguyen, Nitesh V. Chawla, Wray Buntine, Kok-Seng Wong, Khoa D. Doan, Binh T. Nguyen

查看原文 →

PRISM: A Multi-Dimensional Benchmark for Evaluating LLM Peer Reviewers · 相关技术

相关技术

machine learning eval divide-and-conquer partitioning UCT Straight-Through Estimator SSE Prompt NLI LMM LLM-as-a-Judge HIS Grouped Memorization Evaluation Granular Alignment Paradigm For Effort Metric Attention ARG ANN API Mac