LLMs Encode Syntactic Structure Beyond Universal Dependencies, Consistent with Minimalist Phase Theory
A new preprint demonstrates that large language models (LLMs) encode syntactic distinctions not captured by Universal Dependencies (UD) tree distances. Using English wh-movement stimuli, researchers found that probe distances between embedded subjects and verbs reverse sign depending on clause finiteness—a pattern unexplained by UD but consistent with Minimalist phase theory. The findings were robust across 13 models from four families and were validated through causal interventions.
Why it matters: This suggests that LLMs internalize deeper grammatical structures than those represented by standard linguistic annotations, aligning with advanced theoretical linguistics.
Full story at: arXiv Computation and Language ↗