How Well Do Large Language Models Understand Syntax? An Evaluation by Asking Natural Language Questions

TL;DR

This paper evaluates how well large language models understand syntax by using a natural language question-answering approach, revealing limited syntactic comprehension and insights into training dynamics.

Contribution

It introduces a novel question-based evaluation method targeting key syntactic points and provides empirical insights into the syntactic understanding of 24 LLMs.

Findings

Most LLMs show limited syntactic understanding.

Prepositional phrase attachment is particularly challenging.

Syntactic knowledge is mostly acquired early in training.

Abstract

While recent advancements in large language models (LLMs) bring us closer to achieving artificial general intelligence, the question persists: Do LLMs truly understand language, or do they merely mimic comprehension through pattern recognition? This study seeks to explore this question through the lens of syntax, a crucial component of sentence comprehension. Adopting a natural language question-answering (Q&A) scheme, we craft questions targeting nine syntactic knowledge points that are most closely related to sentence comprehension. Experiments conducted on 24 LLMs suggest that most have a limited grasp of syntactic knowledge, exhibiting notable discrepancies across different syntactic knowledge points. In particular, questions involving prepositional phrase attachment pose the greatest challenge, whereas those concerning adjectival modifier and indirect object are relatively easier for LLMs to handle. Furthermore, a case study on the training dynamics of the LLMs reveals that the majority of syntactic knowledge is learned during the initial stages of training, hinting that simply increasing the number of training tokens may not be the `silver bullet' for improving the comprehension ability of LLMs.