大模型能理解句子中谁对谁做了什么，但不如人类敏感。
Do Large Language Models know who did what to whom?
- 通过分析句子表征，发现大模型更关注语法结构而非角色关系。
- 模型在角色反转的句子上相似度判断与人类差异显著。
- 部分注意力头能独立于语法捕捉角色信息，适合角色分析任务。
大语言模型常被批评缺乏语言理解能力，但多数批评针对的是与语言处理无关的认知能力。本文聚焦语言紧密相关的理解——推断句子中的施事、受事等主题角色。大模型的核心训练目标是词预测，这是否导致其句表示包含主题角色信息？我们在四个大模型上进行两项实验，结果表明：与人类相似性判断不同，大模型的整体句表示相似性反映的是句法相似性，而非施事与受事分配是否相同或互换。此外，几乎无证据显示主题角色信息存在于任何隐藏单元子集中。然而，部分注意力头能稳健地捕捉主题角色，且独立于句法。因此，大模型可提取主题角色，但相较于人类，该信息对其表征影响较弱。
原文摘要 · Abstract (English)
Large Language Models (LLMs) are commonly criticized for not understanding language. However, many critiques focus on cognitive abilities that, in humans, are distinct from language processing. Here, we instead study a kind of understanding tightly linked to language: inferring who did what to whom (thematic roles) in a sentence. Does the central training objective of LLMs-word prediction-result in sentence representations that capture thematic roles? In two experiments, we characterized sentence representations in four LLMs. In contrast to human similarity judgments, in LLMs the overall representational similarity of sentence pairs reflected syntactic similarity but not whether their agent and patient assignments were identical vs. reversed. Furthermore, we found little evidence that thematic role information was available in any subset of hidden units. However, some attention heads robustly captured thematic roles, independently of syntax. Therefore, LLMs can extract thematic roles but, relative to humans, this information influences their representations more weakly.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。