Bilingual Lexicon Induction with Semi-supervision in Non-Isometric Embedding Spaces
This work addresses bilingual lexicon induction for natural language processing, offering improved performance especially for etymologically distant languages, but it is incremental as it builds on existing semi-supervised and distribution matching approaches.
The authors tackled the problem of bilingual lexicon induction by relaxing the isometric assumption between embedding spaces, proposing a semi-supervised method called BLISS that leverages limited aligned lexicons and unaligned embeddings. Their method achieved state-of-the-art results on 15 out of 18 language pairs in the MUSE dataset, with particular effectiveness in non-isometric scenarios.
Recent work on bilingual lexicon induction (BLI) has frequently depended either on aligned bilingual lexicons or on distribution matching, often with an assumption about the isometry of the two spaces. We propose a technique to quantitatively estimate this assumption of the isometry between two embedding spaces and empirically show that this assumption weakens as the languages in question become increasingly etymologically distant. We then propose Bilingual Lexicon Induction with Semi-Supervision (BLISS) --- a semi-supervised approach that relaxes the isometric assumption while leveraging both limited aligned bilingual lexicons and a larger set of unaligned word embeddings, as well as a novel hubness filtering technique. Our proposed method obtains state of the art results on 15 of 18 language pairs on the MUSE dataset, and does particularly well when the embedding spaces don't appear to be isometric. In addition, we also show that adding supervision stabilizes the learning procedure, and is effective even with minimal supervision.