We consider the problem of learning general-purpose, paraphrastic sentence embeddings in the setting of Wieting et al. (2016b). We use neural machine translation to generate sentential paraphrases via back-translation of bilingual sentence pairs. We evaluate the paraphrase pairs by their ability to serve as training data for learning paraphrastic sentence embeddings. We find that the data quali...

We describe a set of paraphrase patterns for questions which we derived from a corpus of questions, and report the result of using them in the automatic recognition of question paraphrases. The aim of our paraphrase patterns is to factor out different syntactic variations of interrogative words, since the interrogative part of a question adds a syntactic superstructure on the sentence part (i.e...

This paper proposes a novel method to resolve the coverage problem of SMT system. The method generates paraphrases for source-side sentences of the bilingual parallel data, which are then paired with the target-side sentences to generate new parallel data. Within a statistical paraphrase generation framework, we employ an object function, named Sentence Novelty, to select paraphrases which havi...

We examine the application of data-driven paraphrasing to natural language understanding. We leverage bilingual parallel corpora to extract a large collection of syntactic paraphrase 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 compression task. Further, we use distributional si...

This study evaluates the robustness of established computational indices used to assess text relatedness in userlanguage. The original User-Language Paraphrase Corpus (ULPC) was compared to a corrected version, in which each paraphrase was corrected for typographical and grammatical errors. Error correction significantly affected values for each of five computational indices, indicating greater...

SMT has been used in paraphrase generation by translating a source sentence into another (pivot) language and then back into the source. The resulting sentences can be used as candidate paraphrases of the source sentence. Existing work that uses two independently trained SMT systems cannot directly optimize the paraphrase results. Paraphrase criteria especially the paraphrase rate is not able t...

Most state-of-the-art statistical machine translation systems use log-linear models, which are defined in terms of hypothesis features and weights for those features. It is standard to tune the feature weights in order to maximize a translation quality metric, using heldout test sentences and their corresponding reference translations. However, obtaining reference translations is expensive. In ...

DaNetQA, a new question-answering corpus, follows BoolQ [2] design: it comprises natural yes/no questions. Each question is paired with paragraph from Wikipedia and an answer, derived the paragraph. The task to take both as input come up i.e. produce binary output. In this paper, we present reproducible approach DaNetQA creation investigate transfer learning methods for language transferring. F...