Sparse Lexicalised Features and Topic Adaptation for SMT
Year: 2012
Type: article
Abstract: We present a new approach to domain adaptation for SMT that enriches standard phrase-based models with lexicalised word and phrase pair features to help the model select appro-priate translations for the target domain (TED talks). In addi-tion, we show how source-side sentence-level topics can be incorporated to make the features differentiate between more fine-grained topics within the target domain (topic adapta-tion). We compare tuning our sparse features on a devel-opment set versus on the entire in-domain corpus and intro-duce a new method of porting them to larger mixed-domain models. Experimental results show that our features improve performance over a MIRA baseline and that in some cases we can get additional improvements with topic features. We evaluate our methods on two language pairs, English-French and German-English, showing promising results. 1.
Authors: Eva Hasler, Barry Haddow, Philipp Koehn
Institution: University of Edinburgh
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2293596968