Sparse Bilingual Word Representations for Cross-lingual Lexical Entailment

Proceedings of NAACL-HLT 2016, pages 1187–1197, San Diego, California, June 12-17, 2016. c⃝2016 Association for Computational Linguistics Sparse Bilingual Word Representations for Cross-lingual Lexical Entailment Yogarshi Vyas and Marine Carpuat Department of Computer Science University of Maryland College Park, MD 20742, USA yogarshi@cs.umd.edu, marine@cs.umd.edu Abstract We introduce the task of cross-lingual lexical entailment, which aims to detect whether the meaning of a word in one language can be in- ferred from the meaning of a word in another language. We construct a gold standard for this task, and propose an unsupervised solu- tion based on distributional word representa- tions. As commonly done in the monolingual setting, we assume a word e entails a word f if the prominent context features of e are a sub- set of those of f. To address the challenge of comparing contexts across languages, we pro- pose a novel method for inducing sparse bilin- gual word representations from monolingual and parallel texts. Our approach yields an F- score of 70%, and significantly outperforms strong baselines based on translation and on existing word representations. 1 Introduction Multilingual Natural Language Processing lacks techniques to automatically compare and contrast the meaning of words across languages. Machine translation (Koehn, 2010) lets us discover transla- tion correspondences in bilingual texts, but a word and its translation often do not cover the exact same semantic space: distinct word senses might trans- late differently (Gale et al., 1992; Diab and Resnik, 2002, among others); semantic relations and asso- ciations do not always translate, an important is- sue when constructing multilingual ontologies (Fell- baum and Vossen, 2012); and words in parallel text might be translated non-literally due to lexical gaps (Santos, 1990; Bentivogli and Pianta, 2000) or deci- sions of the translator, as becomes clear when com- paring multiple translations of the same source text (Bhagat and Hovy, 2013). As a result, correct word translations found in par- allel corpora exhibit a variety of relations including equivalence, hypernymy, and meronymy. For in- stance, even after removing noisy examples (John- son et al., 2007) from a Machine Translation bilex- icon induced from parallel corpora (Koehn et al., 2007), we find that the French word “appartement” (apartment) is linked to related but not strictly equiv- alent English words, such as “home” or “condo”. house ||| foyer (foyer) house ||| maison (house) house ||| chambre (chamber) home ||| appartement condo ||| appartement apartment ||| appartement Table 1: Examples of translations drawn from an English- French bilexicon automatically learned on parallel text. We aim to design models that capture these differ- ences and similarities in word meaning across lan- guages, beyond translation correspondences. As a first step, we introduce cross-lingual lexical entail- ment, the task of detecting whether the meaning of a word in one language can be inferred from the meaning of a word in another language. In mono- lingual settings, lexical entailment has received sig- nificant attention as a representation-agnostic way of modeling lexical semantics, and as a step to- ward textual inference (Zhitomirsky-Geffet and …

Content truncated to 500 words