It's not Rocket Science : Interpreting Figurative Language in Narratives

TL;DR

This paper investigates the interpretation of figurative language in narratives, highlighting the challenges for NLP models and proposing knowledge-enhanced approaches that better mimic human understanding.

Contribution

It introduces datasets for idioms and similes in narratives and develops knowledge-enhanced models that improve figurative language interpretation over standard pre-trained models.

Findings

Models outperform baseline on figurative language tasks with knowledge integration

Knowledge-enhanced models close the performance gap with humans

Proposed approach benefits both discriminative and generative tasks

Abstract

Figurative language is ubiquitous in English. Yet, the vast majority of NLP research focuses on literal language. Existing text representations by design rely on compositionality, while figurative language is often non-compositional. In this paper, we study the interpretation of two non-compositional figurative languages (idioms and similes). We collected datasets of fictional narratives containing a figurative expression along with crowd-sourced plausible and implausible continuations relying on the correct interpretation of the expression. We then trained models to choose or generate the plausible continuation. Our experiments show that models based solely on pre-trained language models perform substantially worse than humans on these tasks. We additionally propose knowledge-enhanced models, adopting human strategies for interpreting figurative language types : inferring meaning from the context and relying on the constituent words' literal meanings. The knowledge-enhanced models improve the performance on both the discriminative and generative tasks, further bridging the gap from human performance.