Density Matching for Bilingual Word Embedding

By Chunting Zhou, Xuezhe Ma, Di Wang, Graham Neubig. 2019

41 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Representation & Self-Supervised Learning

Citations in the corpus

Survey of Hallucination in Natural Language Generation (2022, neutral): “Since target and source are two different languages, authors use density matching method for bilingual synonyms from [196].”
Normalizing Flows for Probabilistic Modeling and Inference (2019, background): “Zhou et al. (2019), He et al. (2018), and Jin et al. (2019) define normalizing flows on the continuous space of word embeddings as a subcomponent of models for translation, syntactic structure, and parsing respectively.”
Detecting Hallucinated Content in Conditional Neural Sequence Generation (2020, methodology): “Since it’s not feasible to perform string matching between two languages for MT, we use a bilingual lexicon induction method (Zhou et al., 2019) to ﬁrst translate each English word into a Chinese word and then check its existence in the source text.”
Zero-shot Reading Comprehension by Cross-lingual Transfer Learning with Multi-lingual Language Representation Model (2019, background): “A.2.1 Linear Mapping Method Algorithms proposed in (Lample et al., 2018; Artetxe et al., 2018; Zhou et al., 2019) to unsupervisedly learn linear mapping between two sets of embeddings are used here to align representations of source (training data) to those of target.”
Should All Cross-Lingual Embeddings Speak English? (2019, background): “Starting with a seed dictionary (e.g. from identical strings (Zhou et al., 2019) or numerals) an initial mapping is learned in the same manner as in the supervised setting.”