Abstract
The goal of this paper is to learn more about how idiomatic information is structurally encoded in embeddings, using a structural probing method. We repurpose an existing English verbal multi-word expression (MWE) dataset to suit the probing framework and perform a comparative probing study of static (GloVe) and contextual (BERT) embeddings. Our experiments indicate that both encode some idiomatic information to varying degrees, but yield conflicting evidence as to whether idiomaticity is encoded in the vector norm, leaving this an open question. We also identify some limitations of the used dataset and highlight important directions for future work in improving its suitability for a probing analysis.- Anthology ID:
- 2023.mwe-1.8
- Volume:
- Proceedings of the 19th Workshop on Multiword Expressions (MWE 2023)
- Month:
- May
- Year:
- 2023
- Address:
- Dubrovnik, Croatia
- Venue:
- MWE
- SIG:
- SIGLEX
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 45–57
- Language:
- URL:
- https://aclanthology.org/2023.mwe-1.8
- DOI:
- Cite (ACL):
- Filip Klubička, Vasudevan Nedumpozhimana, and John Kelleher. 2023. Idioms, Probing and Dangerous Things: Towards Structural Probing for Idiomaticity in Vector Space. In Proceedings of the 19th Workshop on Multiword Expressions (MWE 2023), pages 45–57, Dubrovnik, Croatia. Association for Computational Linguistics.
- Cite (Informal):
- Idioms, Probing and Dangerous Things: Towards Structural Probing for Idiomaticity in Vector Space (Klubička et al., MWE 2023)
- PDF:
- https://preview.aclanthology.org/remove-xml-comments/2023.mwe-1.8.pdf