Can Language Models Handle Recursively Nested Grammatical Structures? A Case Study on Comparing Models and Humans
Author:
Open Access
- 1 December 2024
- journal article
- research article
- Published by MIT Press in Computational Linguistics
- Vol. 50 (4) , 1441-1476
Abstract
How should we compare the capabilities of language models (LMs) and humans? In this article, I draw inspiration from comparative psychology to highlight challenges in these comparisons. I focus on a case study: processing of recursively nested grammatical structures. Prior work suggests that LMs cannot process these structures as reliably as humans can. However, the humans were provided with instructions and substantial training, while the LMs were evaluated zero-shot. I therefore match the evaluation more closely. Providing large LMs with a simple prompt—with substantially less content than the human training—allows the LMs to consistently outperform the human results, even in more deeply nested conditions than were tested with humans. Furthermore, the effects of prompting are robust to the particular structures and vocabulary used in the prompt. Finally, reanalyzing the existing human data suggests that the humans may not perform above chance at the difficult structures initially. Thus, large LMs may indeed process recursively nested grammatical structures as reliably as humans, when evaluated comparably. This case study highlights how discrepancies in the evaluation methods can confound comparisons of language models and humans. I conclude by reflecting on the broader challenge of comparing human and model capabilities, and highlight an important difference between evaluating cognitive models and foundation models.Keywords
This publication has 57 references indexed in Scilit:
- Mechanisms for handling nested dependencies in neural-network language models and humansCognition, 2021
- Performance vs. competence in human–machine comparisonsProceedings of the National Academy of Sciences, 2020
- The role of verb repetition in cumulative structural priming in comprehension.Journal of Experimental Psychology: Learning, Memory, and Cognition, 2016
- Recursion: what is it, who has it, and how did it evolve?WIREs Cognitive Science, 2010
- Letting structure emerge: connectionist and dynamical systems approaches to cognitionTrends in Cognitive Sciences, 2010
- The weirdest people in the world?Behavioral and Brain Sciences, 2010
- Assessing the belief bias effect with ROCs: It's a response bias effect.Psychological Review, 2010
- Structural priming: A critical review.Psychological Bulletin, 2008
- Constraints on multiple center-embedding of clausesJournal of Linguistics, 2007
- Becoming syntactic.Psychological Review, 2006