Density Matching for Bilingual Word Embedding
This work addresses the challenge of mapping word embeddings between languages, particularly for difficult language pairs, with potential applications in machine translation and multilingual NLP.
The paper tackled the problem of cross-lingual word embedding by modeling monolingual embedding spaces as probability densities and matching them using normalizing flow, achieving competitive or superior performance on bilingual lexicon induction and cross-lingual word similarity benchmarks, with strong results on distant or morphologically rich languages.
Recent approaches to cross-lingual word embedding have generally been based on linear transformations between the sets of embedding vectors in the two languages. In this paper, we propose an approach that instead expresses the two monolingual embedding spaces as probability densities defined by a Gaussian mixture model, and matches the two densities using a method called normalizing flow. The method requires no explicit supervision, and can be learned with only a seed dictionary of words that have identical strings. We argue that this formulation has several intuitively attractive properties, particularly with the respect to improving robustness and generalization to mappings between difficult language pairs or word pairs. On a benchmark data set of bilingual lexicon induction and cross-lingual word similarity, our approach can achieve competitive or superior performance compared to state-of-the-art published results, with particularly strong results being found on etymologically distant and/or morphologically rich languages.