Semantic Specialisation of Distributional Word Vector Spaces using Monolingual and Cross-Lingual Constraints

Mrkšić, Nikola, and Vulić, Ivan, and Ó Séaghdha, Diarmuid, and Leviant, Ira, and Reichart, Roi, and Gašić, Milica, and Korhonen, Anna, and Young, Steve, (2017) Semantic Specialisation of Distributional Word Vector Spaces using Monolingual and Cross-Lingual Constraints. Transactions of the Association for Computational Linguistics (TACL), 5. pp. 309-324.

Abstract

We present Attract-Repel, an algorithm for improving the semantic quality of word vectors by injecting constraints extracted from lexical resources. Attract-Repel facilitates the use of constraints from mono- and cross-lingual resources, yielding semantically specialised cross-lingual vector spaces. Our evaluation shows that the method can make use of existing cross-lingual lexicons to construct high-quality vector spaces for a plethora of different languages, facilitating semantic transfer from high- to lower-resource ones. The effectiveness of our approach is demonstrated with state-of-the-art results on semantic similarity datasets in six languages. We next show that Attract-Repel-specialised vectors boost performance in the downstream task of dialogue state tracking (DST) across multiple languages. Finally, we show that cross-lingual vector spaces produced by our algorithm facilitate the training of multilingual DST models, which brings further performance improvements.