研究人员推出了 IdioLink，这是一个旨在评估语言模型理解习语能力的新基准。该基准包含超过 10,000 份文档和 2,000 个查询，涵盖 107 个习语，以测试模型是否能将比喻性语言与其概念含义联系起来。当前的嵌入模型在此任务上表现不佳，通常依赖于主题线索而非真正的语义理解，这凸显了在习语感知语义检索方面存在的重大差距。 AI
影响 IdioLink 挑战了当前的语言模型超越字面意义，推动了更深层次的语义理解，并可能提高了 AI 对细微差别的人类语言的掌握能力。
排序理由 该集群包含一篇介绍用于评估语言模型的新基准的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2605.22247v1 Announce Type: new Abstract: Idioms pose a fundamental challenge for language models, as their meaning cannot be inferred from surface form alone. Understanding such expressions, therefore, requires semantic abstraction beyond lexical overlap. We introduce Idio…
Idioms pose a fundamental challenge for language models, as their meaning cannot be inferred from surface form alone. Understanding such expressions, therefore, requires semantic abstraction beyond lexical overlap. We introduce IdioLink, a retrieval benchmark designed to test whe…