Conference paper Open Access
Xu, Jitao; Crego, Josep; Senellart, Jean
<?xml version='1.0' encoding='utf-8'?> <oai_dc:dc xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd"> <dc:creator>Xu, Jitao</dc:creator> <dc:creator>Crego, Josep</dc:creator> <dc:creator>Senellart, Jean</dc:creator> <dc:date>2019-11-02</dc:date> <dc:description>This work is inspired by a typical machine translation industry scenario in which translators make use of in-domain data for facilitating translation of similar or repeating sentences. We introduce a generic framework applied at inference in which a subset of segment pairs are first extracted from training data according to their similarity to the input sentences. These segments are then used to dynamically update the parameters of a generic NMT network, thus performing a lexical micro-adaptation. Our approach demonstrates strong adaptation performance to new and existing datasets including pseudo in-domain data. We evaluate our approach on a heterogeneous English-French training dataset showing accuracy gains on all evaluated domains when compared to strong adaptation baselines.</dc:description> <dc:identifier>https://zenodo.org/record/3524977</dc:identifier> <dc:identifier>10.5281/zenodo.3524977</dc:identifier> <dc:identifier>oai:zenodo.org:3524977</dc:identifier> <dc:language>eng</dc:language> <dc:relation>doi:10.5281/zenodo.3524976</dc:relation> <dc:relation>url:https://zenodo.org/communities/iwslt2019</dc:relation> <dc:rights>info:eu-repo/semantics/openAccess</dc:rights> <dc:rights>https://creativecommons.org/licenses/by/4.0/legalcode</dc:rights> <dc:title>Lexical Micro-adaptation for Neural Machine Translation</dc:title> <dc:type>info:eu-repo/semantics/conferencePaper</dc:type> <dc:type>publication-conferencepaper</dc:type> </oai_dc:dc>
All versions | This version | |
---|---|---|
Views | 129 | 129 |
Downloads | 99 | 99 |
Data volume | 42.8 MB | 42.8 MB |
Unique views | 109 | 109 |
Unique downloads | 86 | 86 |