Word Definitions from Large Language Models
This addresses the problem of evaluating computational word definitions for linguists and NLP practitioners, but it is incremental as it builds on existing generative models.
The study compared word definitions from traditional dictionaries with those generated by ChatGPT variants, finding that ChatGPT definitions are highly accurate (comparable to dictionaries) and perform better than GloVE and FastText embeddings on low-frequency words.
Dictionary definitions are historically the arbitrator of what words mean, but this primacy has come under threat by recent progress in NLP, including word embeddings and generative models like ChatGPT. We present an exploratory study of the degree of alignment between word definitions from classical dictionaries and these newer computational artifacts. Specifically, we compare definitions from three published dictionaries to those generated from variants of ChatGPT. We show that (i) definitions from different traditional dictionaries exhibit more surface form similarity than do model-generated definitions, (ii) that the ChatGPT definitions are highly accurate, comparable to traditional dictionaries, and (iii) ChatGPT-based embedding definitions retain their accuracy even on low frequency words, much better than GloVE and FastText word embeddings.