Learned in Translation: Contextualized Word Vectors

TL;DR

This paper introduces CoVe, contextualized word vectors derived from a machine translation encoder, which significantly enhance NLP task performance over traditional static embeddings.

Contribution

The paper presents a novel method to generate contextualized word vectors from a translation model encoder, improving NLP task results beyond existing static embeddings.

Findings

CoVe improves sentiment analysis accuracy.

CoVe enhances entailment task performance.

State-of-the-art results achieved on fine-grained tasks.

Abstract

Computer vision has benefited from initializing multiple deep layers with weights pretrained on large supervised training sets like ImageNet. Natural language processing (NLP) typically sees initialization of only the lowest layer of deep models with pretrained word vectors. In this paper, we use a deep LSTM encoder from an attentional sequence-to-sequence model trained for machine translation (MT) to contextualize word vectors. We show that adding these context vectors (CoVe) improves performance over using only unsupervised word and character vectors on a wide variety of common NLP tasks: sentiment analysis (SST, IMDb), question classification (TREC), entailment (SNLI), and question answering (SQuAD). For fine-grained sentiment analysis and entailment, CoVe improves performance of our baseline models to the state of the art.