Probing Multilingual Cognate Prediction Models

Clémentine Fourrier, Benoît Sagot


Abstract
Character-based neural machine translation models have become the reference models for cognate prediction, a historical linguistics task. So far, all linguistic interpretations about latent information captured by such models have been based on external analysis (accuracy, raw results, errors). In this paper, we investigate what probing can tell us about both models and previous interpretations, and learn that though our models store linguistic and diachronic information, they do not achieve it in previously assumed ways.
Anthology ID:
2022.findings-acl.299
Volume:
Findings of the Association for Computational Linguistics: ACL 2022
Month:
May
Year:
2022
Address:
Dublin, Ireland
Editors:
Smaranda Muresan, Preslav Nakov, Aline Villavicencio
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3786–3801
Language:
URL:
https://aclanthology.org/2022.findings-acl.299
DOI:
10.18653/v1/2022.findings-acl.299
Bibkey:
Cite (ACL):
Clémentine Fourrier and Benoît Sagot. 2022. Probing Multilingual Cognate Prediction Models. In Findings of the Association for Computational Linguistics: ACL 2022, pages 3786–3801, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Probing Multilingual Cognate Prediction Models (Fourrier & Sagot, Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-acl.299.pdf
Video:
 https://aclanthology.org/2022.findings-acl.299.mp4