Monolingual Machine Translation for Paraphrase Generation

By Chris Quirk, Chris Brockett, William B. Dolan. Conference on Empirical Methods in Natural Language Processing, 2004

359 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/2a5a179a70e98ee1c4b4b8b54f6c8aba629c02e4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Data-Driven Response Generation in Social Media (2011, background): “…(Wong and Mooney, 2006; Wong and Mooney, 2007), summarization (Daum´e III and Marcu, 2009), generating bid-phrases in on-line advertising (Ravi et al., 2010), spelling correction (Sun et al., 2010), paraphrase (Dolan et al., 2004; Quirk et al., 2004) and query expansion (Rie-zler et al., 2007).”
ParaNMT-50M: Pushing the Limits of Paraphrastic Sentence Embeddings with Millions of Machine Translations (2017, methodology): “Many methods have been developed for generating or finding paraphrases, including using multiple translations of the same source material (Barzilay and McKeown, 2001), using distributional similarity to find similar dependency paths (Lin and Pantel, 2001), using comparable articles from multiple new”
Generating Phrasal and Sentential Paraphrases: A Survey of Data-Driven Methods (2010, methodology): “Sentential paraphrases may be evaluated in a similar fashion without the need for any surrounding context (Quirk, Brockett, and Dolan 2004).”
PPDB: The Paraphrase Database (2013, methodology): “Past paraphrase collections include automatically derived resources like DIRT (Lin and Pantel, 2001), the MSR paraphrase corpus and phrase table (Dolan et al., 2004; Quirk et al., 2004), among others.”
Unsupervised Paraphrase Generation using Pre-trained Language Models (2020, background): “Some earlier approaches towards paraphrasing like [3], [4] and [5] were mostly based on linguistic structures.”
QuickEdit: Editing Text & Translations by Crossing Words Out (2017, methodology): “paraphrasing (Quirk et al., 2004), or a translation, i.”
Retrieving and Reading: A Comprehensive Survey on Open-domain Question Answering (2021, methodology): “To address this problem, query expansion [48], [49] and paraphrasing techniques [50], [51], [52], [53] are often employed to produce additional search words or phrases so as to retrieve more relevant documents.”
CRUISE: Cold-Start New Skill Development via Iterative Utterance Generation (2018, background): “One class of approaches is to paraphrase user utterances to increase the number of training set (Barzilay and Lee, 2003; Quirk et al., 2004; Kauchak and Barzilay, 2006; Zhao et al., 2009; Prakash et al., 2016).”
Deep Learning for Joint Source-Channel Coding of Text (2018, background): “it distance normalized by the length of the sentence is what we refer to as the word error rate. Word error rate is commonly used to evaluate performance in speech recognition and machine translation [30, 31]. A downside of the metric is that it cannot capture the effect of synonyms or other aspects”
Paraphrase-Driven Learning for Open Question Answering (2013, methodology): “The learning algorithms presented in this paper are similar to algorithms used for paraphrase extraction from sentence-aligned corpora (Barzi-lay and McKeown, 2001; Barzilay and Lee, 2003; Quirk et al., 2004; Bannard and Callison-Burch, 2005; Callison-Burch, 2008; Marton et al., 2009).”
Evaluation of Text Generation: A Survey (2020, methodology): “There is a large body of work that is based on template-based models and have used statistical methods to improve generation by introducing new methods such as sentence compression, reordering, lexical paraphrasing, and syntactic transformation, to name a few (Sporleder, 2005; Steinberger, 2006; Kni”
Sentence Compression Beyond Word Deletion (2008, background): “Although there is a greater supply of paraphrasing corpora, such as the Multiple-Translation Chinese (MTC) corpus 1 and the Microsoft Research (MSR) Paraphrase Corpus (Quirk et al., 2004), they are also not ideal, since they have not been created with compression in mind.”
QuickEdit: Editing Text & Translations via Simple Delete Actions (2017, methodology): “Specifically, we consider the task of reformulating either a sentence, i.e. paraphrasing (Quirk et al., 2004), or a translation, i.e. translation postediting (Koehn, 2009b).”
Split and Rephrase (2017, background): “…machine translation (SMT) based systems which learn paraphrases from monolingual parallel (Barzilay and McKeown, 2001; Zhao et al., 2008), comparable (Quirk et al., 2004) or bilingual parallel (Bannard and Callison-Burch, 2005; Gan-itkevitch et al., 2011) corpora; and a recent neural machine…”
Beyond SumBasic: Task-focused summarization with sentence simplification and lexical expansion (2007, unknown): “The reader is referred to Quirk et al, (2004) and Dolan et al., (2005) for further information about the construction of this corpus.”
Paraphrase Identification as Probabilistic Quasi-Synchronous Recognition (2009, methodology): “In all our experiments, we have used the Microsoft Research Paraphrase Corpus (Dolan et al., 2004; Quirk et al., 2004).”
Adversarial Example Generation with Syntactically Controlled Paraphrase Networks (2018, methodology): “The second family includes methods that rewrite the input using methods based on parallel text (Bannard and Callison-Burch, 2005), machine translation (Quirk et al., 2004; Napoles et al., 2016; Suzuki et al., 2017), or related statistical techniques (Zhao et al.”
From Paraphrase Database to Compositional Paraphrase Model and Back (2015, background): “, 2014b), or entire sentences, such as the Microsoft Research Paraphrase Corpus (Dolan et al., 2004; Quirk et al., 2004).”