arXiv cs.CL· H\'ector Javier V\'azquez Mart\'inez, Charles Yang·· 11 小时前AI 评分34
用儿童语言习得指标衡量语言模型的形式与功能能力
Measuring Form and Function in Language Models
AI 导读
研究提出 Contextual Alternative Choice（CAC）提示方法，用儿童语言习得中关于英语限定词的形式句法与功能语篇指标来评测语言模型。结果显示，在可比数据量下训练的现有模型均无法同时达到人类儿童的形式与功能基准，但部分超大规模模型可以做到。
正文 · 原文
Abstract:We introduce quantitative metrics for child language acquisition to evaluate language models. Our focus is on the formal syntactic and functional discourse properties of determiners in English, which young children acquire early and accurately. We propose Contextual Alternative Choice (CAC), a new prompting method which provides targeted tests for both syntactic and discourse knowledge of language. The method enables direct comparison of language models against children, and more importantly, against statistical benchmarks independently established in empirical research. No current model trained on a comparable amount of data simultaneously meet both formal and functional benchmarks like human children, but some very large models do. We present our results as methodological and technical contributions, with specific emphasis on cognitive status of language models.
| Comments: | Under review at ACL Rolling Review May 2026 cycle |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.28616 [cs.CL] |
| (or arXiv:2605.28616v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2605.28616 arXiv-issued DOI via DataCite |
|
| Journal reference: | 2026.babylm-main, pages 438-458, Budapest, Hungary. Association for Computational Linguistics |
| Related DOI: | https://doi.org/10.18653/2026.babylm-main.35
DOI(s) linking to related resources |
Submission history
From: Héctor Vázquez Martínez [view email]
[v1]
Wed, 27 May 2026 15:27:16 UTC (3,033 KB)
[v2]
Thu, 5 Nov 2026 21:05:28 UTC (3,033 KB)
来源：arXiv cs.CL · arxiv.org