What does it mean to be language-agnostic? Probing multilingual sentence encoders for typological properties
This work addresses the need for interpretability in multilingual NLP models, though it is incremental as it builds on existing encoders without introducing new methods.
The paper tackled the problem of understanding what typological properties multilingual sentence encoders capture across languages, and found that different pretraining strategies lead to variations in how linguistic features are encoded in model layers.
Multilingual sentence encoders have seen much success in cross-lingual model transfer for downstream NLP tasks. Yet, we know relatively little about the properties of individual languages or the general patterns of linguistic variation that they encode. We propose methods for probing sentence representations from state-of-the-art multilingual encoders (LASER, M-BERT, XLM and XLM-R) with respect to a range of typological properties pertaining to lexical, morphological and syntactic structure. In addition, we investigate how this information is distributed across all layers of the models. Our results show interesting differences in encoding linguistic variation associated with different pretraining strategies.