The Holistic Storage of Verb+Up Phrases in Text-based and Audio-based Language Models
For NLP researchers, it provides evidence that language models exhibit usage-based storage patterns, but the findings are incremental as they extend known phenomena to new models.
This study investigates whether text-based LLMs and an ASR model store multi-word units (V+up phrasal verbs) holistically, finding that frequency and predictability drive holistic storage, supporting usage-based linguistic theories.
A crucial aspect of linguistic capability is the ability to trade off between stored representations and abstract knowledge: one must retrieve learned representations, but also generate novel ones by applying productive rules. While recent work has examined abstract knowledge in language models, holistic storage of multi-word units has received far less attention. We probe internal representations in text-based LLMs and an ASR model, testing whether V+up phrasal verbs develop distinct representations as a function of frequency and predictability. All models show evidence of holistic storage driven by frequency and predictability, further supporting usage-based theories of language.