大模型虽未显式学习语法，却能识别复杂句法结构。
Grammaticality Judgments in Humans and Language Models: Revisiting Generative Grammar with LLMs
- 通过表面文本训练，模型自发产生对句法结构的敏感性。
- 在主谓倒装和寄生空位中，模型准确区分合语法与不合语法句子。
- 适合关注语言模型认知机制的研究者阅读。
什么可作为句法结构的证据？传统生成语法认为，主谓倒装和寄生空位等系统性语法判断是内部层级语法存在的证据。本文测试大语言模型（LLMs）是否仅通过表面形式训练，也能再现这些对比，从而推断其具备潜在结构表征。我们聚焦两类经典句法结构：主谓倒装（检验对主语边界的识别能力）和寄生空位许可（检验抽象依存结构）。采用提示词诱发接受度评分，评估GPT-4和LLaMA-3等模型。结果显示，模型在两类构造中均能可靠区分合语法与不合语法变体，表明其对句法结构敏感，而不仅是线性顺序。预测性训练使模型自发产生超越认知知识的结构归纳能力，暗示语法敏感性可在无显式编码下实现。
原文摘要 · Abstract (English)
What counts as evidence for syntactic structure? In traditional generative grammar, systematic contrasts in grammaticality such as subject-auxiliary inversion and the licensing of parasitic gaps are taken as evidence for an internal, hierarchical grammar. In this paper, we test whether large language models (LLMs), trained only on surface forms, reproduce these contrasts in ways that imply an underlying structural representation. We focus on two classic constructions: subject-auxiliary inversion (testing recognition of the subject boundary) and parasitic gap licensing (testing abstract dependency structure). We evaluate models including GPT-4 and LLaMA-3 using prompts eliciting acceptability ratings. Results show that LLMs reliably distinguish between grammatical and ungrammatical variants in both constructions, and as such support that they are sensitive to structure and not just linear order. Structural generalizations, distinct from cognitive knowledge, emerge from predictive training on surface forms, suggesting functional sensitivity to syntax without explicit encoding.
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。