Unsupervised Paraphrasing of Multiword Expressions

We propose an unsupervised approach to paraphrasing multiword expressions (MWEs) in context. Our model employs only monolingual corpus data and pre-trained language models (without fine-tuning), and does not make use of any external resources such as dictionaries. We evaluate our method on the SemEval 2022 idiomatic semantic text similarity task, and show that it outperforms all unsupervised systems and rivals supervised systems. 1 * This work was partially done when the first author was at Riken. 1 Code is available at: https://github.com/ twadada/mwe-paraphrase . 2 The MWE is said to originate from an ancient legend that a swan sings beautifully before it dies.