For a Fistful of Puns: Evaluating a Puns in Multiword Expressions Identification Algorithm Without Dedicated Dataset
Julien Bezançon, Gaël Lejeune · 2025
Machine Translation systems have always struggled with challenges such as multiword expressions (MWEs) and wordplays.These phenomena are idiosyncratic, pervasive across languages, and significantly affect performance of MT systems (among others).In this context, we explore the nature of puns created from multiword expressions (PMWEs), characterized by the creation of a wordplay from a source MWE to recontextualize it or to give it a humorous touch.Little work has been done on PMWEs in NLP.To address this challenge, we introduce ASMR, an alignment-based PMWE identification and tagging algorithm.We offer an in-depth analysis of three different approaches to ASMR, each created to identify different types of PMWEs.In the absence of PMWErelated datasets and resources, we proceed to a snowclone detection task in English.We also perform a MWE identification task in 26 languages to evaluate ASMR performance across different languages.We show that ASMR exhibits state-of-the-art results for the snowclone detection task and produces interesting results with the MWE identification task.These results may indicate that ASMR is suitable for a PMWE identification task.