Paper: Optimal and Syntactically-Informed Decoding for Monolingual Phrase-Based Alignment

ACL ID P11-2044
Title Optimal and Syntactically-Informed Decoding for Monolingual Phrase-Based Alignment
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2011
Authors

The task of aligning corresponding phrases across two related sentences is an important component of approaches for natural language problems such as textual inference, paraphrase detection and text-to-text generation. In this work, we examine a state-of-the-art struc- tured prediction model for the alignment task which uses a phrase-based representation and is forced to decode alignments using an ap- proximate search approach. We propose in- stead a straightforward exact decoding tech- nique based on integer linear programming that yields order-of-magnitude improvements in decoding speed. This ILP-based decoding strategy permits us to consider syntactically- informed constraints on alignments which sig- nificantly increase the precision of the model.