Same Neurons, Different Languages: Probing Morphosyntax in Multilingual Pre-trained Models
Karolina Stanczak, Edoardo Ponti, Lucas Torroba Hennigen, Ryan Cotterell, Isabelle Augenstein
Abstract
The success of multilingual pre-trained models is underpinned by their ability to learn representations shared by multiple languages even in absence of any explicit supervision. However, it remains unclear how these models learn to generalise across languages. In this work, we conjecture that multilingual pre-trained models can derive language-universal abstractions about grammar. In particular, we investigate whether morphosyntactic information is encoded in the same subset of neurons in different languages. We conduct the first large-scale empirical study over 43 languages and 14 morphosyntactic categories with a state-of-the-art neuron-level probe. Our findings show that the cross-lingual overlap between neurons is significant, but its extent may vary across categories and depends on language proximity and pre-training data size.- Anthology ID:
- 2022.naacl-main.114
- Volume:
- Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
- Month:
- July
- Year:
- 2022
- Address:
- Seattle, United States
- Venue:
- NAACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 1589–1598
- Language:
- URL:
- https://aclanthology.org/2022.naacl-main.114
- DOI:
- 10.18653/v1/2022.naacl-main.114
- Cite (ACL):
- Karolina Stanczak, Edoardo Ponti, Lucas Torroba Hennigen, Ryan Cotterell, and Isabelle Augenstein. 2022. Same Neurons, Different Languages: Probing Morphosyntax in Multilingual Pre-trained Models. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 1589–1598, Seattle, United States. Association for Computational Linguistics.
- Cite (Informal):
- Same Neurons, Different Languages: Probing Morphosyntax in Multilingual Pre-trained Models (Stanczak et al., NAACL 2022)
- PDF:
- https://preview.aclanthology.org/auto-file-uploads/2022.naacl-main.114.pdf
- Code
- copenlu/multilingual-typology-probing