In this work, we carry out two experiments in order to assess the ability of BERT to capture themeaning shift associated with metonymic expressions. We test the model on a new dataset that isrepresentative of the most common types of metonymy. We compare BERT with the StructuredDistributional Model (SDM), a model for the representation of words in context which is basedon the notion of Generalized Event Knowledge. The results reveal that, while BERT abilityto deal with metonymy is quite limited, SDM is good at predicting the meaning of metonymicexpressions, providing support for an account of metonymy based on event knowledge.

Don’t Invite BERT to Drink a Bottle: Modeling the Interpretation of Metonymies Using BERT and Distributional Representations

Paolo Pedinotti
Primo
;
Alessandro Lenci
Secondo
2020-01-01

Abstract

In this work, we carry out two experiments in order to assess the ability of BERT to capture themeaning shift associated with metonymic expressions. We test the model on a new dataset that isrepresentative of the most common types of metonymy. We compare BERT with the StructuredDistributional Model (SDM), a model for the representation of words in context which is basedon the notion of Generalized Event Knowledge. The results reveal that, while BERT abilityto deal with metonymy is quite limited, SDM is good at predicting the meaning of metonymicexpressions, providing support for an account of metonymy based on event knowledge.
2020
978-1-952148-27-9
File in questo prodotto:
File Dimensione Formato  
Pedinotti_Lenci_COLING2020.pdf

accesso aperto

Descrizione: Articolo principale
Tipologia: Versione finale editoriale
Licenza: Creative commons
Dimensione 179.57 kB
Formato Adobe PDF
179.57 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11568/1069937
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact