Rolling the DICE on Idiomaticity: How LLMs Fail to Grasp Context

TL;DR

This paper investigates the ability of large language models to understand idiomatic expressions in context, revealing their limitations in disambiguating idiomatic meaning despite high performance on standard tasks.

Contribution

The authors create a novel contrastive dataset to specifically test LLMs' contextual understanding of idioms and analyze factors like frequency and sentence probability affecting performance.

Findings

LLMs often fail to resolve idiomaticity when context is necessary.

Model performance improves on sentences with higher likelihood.

Collocational frequency influences model accuracy.

Abstract

Human processing of idioms relies on understanding the contextual sentences in which idioms occur, as well as language-intrinsic features such as frequency and speaker-intrinsic factors like familiarity. While LLMs have shown high performance on idiomaticity detection tasks, this success may be attributed to reasoning shortcuts in existing datasets. To this end, we construct a novel, controlled contrastive dataset designed to test whether LLMs can effectively use context to disambiguate idiomatic meaning. Additionally, we explore how collocational frequency and sentence probability influence model performance. Our findings reveal that LLMs often fail to resolve idiomaticity when it is required to attend to the surrounding context, and that models perform better on sentences that have higher likelihood. The collocational frequency of expressions also impacts performance. We make our code and dataset publicly available.