Conference paper Open Access

Lexical Micro-adaptation for Neural Machine Translation

Xu, Jitao; Crego, Josep; Senellart, Jean

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.

Files (432.3 kB)
Name Size
IWSLT2019_paper_9.pdf
md5:b145694f9a8fd725c74837681d966e22
432.3 kB Download
125
96
views
downloads
All versions This version
Views 125125
Downloads 9696
Data volume 41.5 MB41.5 MB
Unique views 105105
Unique downloads 8484

Share

Cite as