Paper: Large-Scale Paraphrasing for Natural Language Understanding

ACL ID N13-2009
Title Large-Scale Paraphrasing for Natural Language Understanding
Venue Annual Conference of the North American Chapter of the Association for Computational Linguistics
Session Student Session
Year 2013
Authors

We examine the application of data-driven paraphrasing to natural language understand- ing. We leverage bilingual parallel corpora to extract a large collection of syntactic para- phrase pairs, and introduce an adaptation scheme that allows us to tackle a variety of text transformation tasks via paraphrasing. We evaluate our system on the sentence compres- sion task. Further, we use distributional sim- ilarity measures based on context vectors de- rived from large monolingual corpora to anno- tate our paraphrases with an orthogonal source of information. This yields significant im- provements in our compression system?s out- put quality, achieving state-of-the-art perfor- mance. Finally, we propose a refinement of our paraphrases by classifying them into nat- ural logic entailment relation...