Learning Bilingual Word Representations by Marginalizing Alignments

Abstract

We present a probabilistic model that simultaneously learns alignments and distributed representations for bilingual data. By marginalizing over word alignments the model captures a larger semantic context than prior work relying on hard alignments. The advantage of this approach is demonstrated in a cross-lingual classification task, where we outperform the prior published state of the art.

Proceedings of ACL 2014 (Short Papers)

References in corpus (4)

• Playing Atari with Deep Reinforcement Learning
• Exploiting Similarities among Languages for Machine Translation
• Multilingual Models for Compositional Distributed Semantics
• Learning Multilingual Word Representations using a Bag-of-Words Autoencoder

Cited by in corpus (11)

• Massively Multilingual Word Embeddings
• XNLI: Evaluating Cross-lingual Sentence Representations
• Separated by an Un-common Language: Towards Judgment Language Informed Vector Space Modeling
• A Novel Bilingual Word Embedding Method for Lexical Translation Using Bilingual Sense Clique
• Cross-Lingual BERT Transformation for Zero-Shot Dependency Parsing
• CoSDA-ML: Multi-Lingual Code-Switching Data Augmentation for Zero-Shot Cross-Lingual NLP
• Cross-lingual Alignment vs Joint Training: A Comparative Study and A Simple Unified Framework
• A Cross-Architecture Instruction Embedding Model for Natural Language Processing-Inspired Binary Code Analysis
• Characterizing Departures from Linearity in Word Translation
• CLUSE: Cross-Lingual Unsupervised Sense Embeddings
• Adversarial Structured Prediction for Multivariate Measures