Evaluating a Multi-sense Definition Generation Model for Multiple Languages

TL;DR

This paper introduces a context-agnostic multi-sense definition generation model that produces multiple definitions for polysemous words across nine languages, outperforming single-sense models on diverse datasets.

Contribution

The study presents a novel multi-sense, language-agnostic definition modeling approach evaluated on fifteen datasets, extending beyond English and addressing polysemy.

Findings

Multi-sense model outperforms single-sense model on all datasets

Effective across nine languages and fifteen datasets

Uses variations of BLEU for evaluation

Abstract

Most prior work on definition modeling has not accounted for polysemy, or has done so by considering definition modeling for a target word in a given context. In contrast, in this study, we propose a context-agnostic approach to definition modeling, based on multi-sense word embeddings, that is capable of generating multiple definitions for a target word. In further, contrast to most prior work, which has primarily focused on English, we evaluate our proposed approach on fifteen different datasets covering nine languages from several language families. To evaluate our approach we consider several variations of BLEU. Our results demonstrate that our proposed multi-sense model outperforms a single-sense model on all fifteen datasets.