Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation

Обучение представлений фраз с использованием RNN энкодер–декодера для статистического машинного перевода
Çağlar Gülçehre, Yoshua Bengio, Kyunghyun Cho, Bart van Merriënboer, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk
2014-01-01
SCID:  54.1/9qm76zut Discuss with AI
EMNLP 2014RNN Encoder–Decoderphrase representationssequence-to-sequencestatistical machine translation
Abstract (AI)
Kyunghyun Cho, Bart van Merriënboer, Caglar Gulcehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, Yoshua Bengio. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2014.
Key Findings
1
Demonstrates that scoring phrase pairs with the RNN Encoder–Decoder improves translation quality when integrated into a statistical MT system.
2
Introduces an RNN Encoder–Decoder architecture to learn variable-length phrase representations for statistical machine translation.
3
Shows the architecture can handle variable-length input and output without relying on a fixed vocabulary alignment.
4
The model jointly learns to encode a source phrase into a fixed-length vector and decode it into a target phrase.
Research Object

RNN Encoder–Decoder model for learning phrase representations in statistical machine translation

Research Subject

Learning phrase representations (vector embeddings) and their use to improve statistical machine translation performance

Publication Details
Publication Date
2014-01-01
Journal
Publisher
ISSN
Access Type
Author Information
Authors
Çağlar Gülçehre
Yoshua Bengio
Kyunghyun Cho
Bart van Merriënboer
Dzmitry Bahdanau
Fethi Bougares
Holger Schwenk
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%