作者位次 #2/5ACL 2026Poster· Natural Language Processing
Quantifying Metric and Model Agreement in Bias Evaluation of Large Language Models
Arash Asgari · Huan Wu · Amirreza Naziri · Mojtaba Kolahdouzi · Laleh Seyyed-Kalantari
摘要
Bias evaluation in large language models (LLMs) uses many metrics and benchmarks, but lacks a systematic way to measure agreement across bias metrics and models. As a result, improvements observed under one metric may contradict another, and model rankings may reflect benchmark-specific artifacts...