arXiv cs.CL· Mervat Abassy, Zhaoxin Feng, Tatsuki Kuribayashi, Kentaro Inui, Emmanuele Chersoni·· 12 小时前AI 评分29
为什么 LLM 会接受泛指陈述？用心理语言学和模型衍生因素解释可接受性
Why Do LLMs Accept Generics? Explaining Acceptability using Psycholinguistic and Model-Derived Factors
AI 导读
研究用心理语言学数据集和自然语料测试 12 个模型，发现 LLM 判断泛指陈述时会像人一样权衡属性常见度与区分度，并偏好特征性属性，但忽略显著性属性。对 LLM 影响最大的是陈述是否事实为真，人类因素的多数影响通过这一判断传导。在 6 个 7-9B 模型上，沿各因素方向移动隐藏状态可让判断按该因素预测的方向变化。
正文 · 原文
Abstract:Psycholinguistic studies explore why humans accept some generic statements but reject others by showing that these judgments depend on more than how common a property is. People consider several other factors such as how well it sets the kind apart from others or whether it is characteristic of the kind.
Inspired by the linguistic and cognitive literature, we explore to what extent factors from these studies explain how large language models (LLMs) judge generics. Our results with twelve models on a psycholinguistic dataset and on naturally occurring generics show that models weigh how common a property is against how well it distinguishes the kind, and combine the two much as people do, though less clearly. Like people, they favor characteristic properties, but unlike the accounts of human judgment, they ignore striking ones. What matters most to LLMs is whether a statement is factually true, and much of the influence of human factors reaches their ratings through that judgment. In six 7-9B models, shifting the hidden states along each factor's direction moves the judgment as the factor predicts.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.27928 [cs.CL] |
| (or arXiv:2610.27928v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.27928 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Mervat Abassy [view email]
[v1]
Tue, 20 Oct 2026 09:40:13 UTC (270 KB)
来源：arXiv cs.CL · arxiv.org