arXiv cs.CL· Michael Wilson, Robert Frank·· 17 小时前AI 评分31
研究对比人类与大语言模型基于形式信息的语言能产性
Examining form-based linguistic productivity in people and large language models
AI 导读
一项提交至 Nature Human Behaviour 的研究对比了人类与大语言模型在基于形式信息的语言能产性上的差异。人类能将主动语态中学到的新动词推广到被动语态，即便该动词无法赋予合理含义；而大语言模型只有在能为新动词赋予类似已有动词的含义时才能成功。研究据此认为，计算模型的句法知识仍与具体词汇项的分布和含义绑定，人类的句法知识则不然。
正文 · 原文
Abstract:Linguistic productivity, as evidenced by the use of words in novel contexts, is one of the most striking facts about human language. However, how abstract the knowledge that enables such productivity is remains unclear. We compare people's and computational language models' linguistic productivity in contexts where successful generalization depends on form-based information. We focus on extending the use of a novel verb learned in the active voice to passive contexts. We find people extend active to passive uses in a way consistent with abstract generalizations concerning linguistic form, even when no plausible meaning can be assigned to the novel verb. In contrast, large language models succeeded only when a meaning like an existing verb's could be assigned to the novel verb. We conclude that computational models' syntactic knowledge remains tied to the distribution and meaning of specific lexical items in a way that human syntactic knowledge is not.
| Comments: | Submitted to Nature Human Behaviour |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.33433 [cs.CL] |
| (or arXiv:2610.33433v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33433 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Michael Wilson [view email]
[v1]
Sat, 24 Oct 2026 04:21:51 UTC (235 KB)
来源：arXiv cs.CL · arxiv.org