JuICE: A Benchmark for Evaluating LLM-Judge in Identifying Cultural Errors 文章

ArXiv CS.CL2026-05-27NEWSen作者: Jiho Jin, Junho Myung, Juhyun Oh, Junyeong Park, Rifki Afina Putri, Sunipa Dev, Vinodkumar Prabhakaran, Alice Oh

大语言模型

查看原文 →

JuICE: A Benchmark for Evaluating LLM-Judge in Identifying Cultural Errors · 相关技术

相关技术

ORM ODE PLA LLM language model 研究法辛烷值 eval divide-and-conquer partitioning Verse SPA PPR OTA NAT LMM large language models LLM-as-a-Judge HIS Granular Alignment Paradigm For ARG ANN ICE RAW