Multilingual Multiword Expression Identification Using Lateral Inhibition and Domain Adaptation
This work addresses the challenge of accurately identifying multiword expressions for natural language processing systems across multiple languages, representing an incremental improvement over existing methods.
The paper tackled the problem of multilingual multiword expression identification by training mBERT on 14 languages from the PARSEME corpus, incorporating lateral inhibition and adversarial training, and achieved better results than the best competition system on most languages, with average improvements of 1.23% and 4.73% on global and unseen MWE identification.
Correctly identifying multiword expressions (MWEs) is an important task for most natural language processing systems since their misidentification can result in ambiguity and misunderstanding of the underlying text. In this work, we evaluate the performance of the mBERT model for MWE identification in a multilingual context by training it on all 14 languages available in version 1.2 of the PARSEME corpus. We also incorporate lateral inhibition and language adversarial training into our methodology to create language-independent embeddings and improve its capabilities in identifying multiword expressions. The evaluation of our models shows that the approach employed in this work achieves better results compared to the best system of the PARSEME 1.2 competition, MTLB-STRUCT, on 11 out of 14 languages for global MWE identification and on 12 out of 14 languages for unseen MWE identification. Additionally, averaged across all languages, our best approach outperforms the MTLB-STRUCT system by 1.23% on global MWE identification and by 4.73% on unseen global MWE identification.