NLP Wiki

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== Paraphrase ======
A common linguistic definition of paraphrase is that two sentences are paraphrases of each other if they mutually entail each other.

===== Paraphrase Identification =====
  * [[https://arxiv.org/pdf/1711.05732.pdf|Wieting & Gimpel 2017 - ParaNMT-50M: Pushing the Limits of Paraphrastic Sentence Embeddings with Millions of Machine Translations]]

==== Datasets ====
  * Microsoft Research Paraphrase Corpus: [[https://www.microsoft.com/en-us/download/details.aspx?id=52398|Dataset]], paper: [[https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/I05-50025B15D.pdf|Dolan & Brockett 2003 - Automatically Constructing a Corpus of Sentential Paraphrases]]
  * PAWS: [[https://arxiv.org/pdf/1904.01130.pdf|Zhang et al 2019 - PAWS: Paraphrase Adversaries from Word Scrambling]]
  * PAWS-X: [[https://arxiv.org/pdf/1908.11828.pdf|Yang et al 2019 - PAWS-X: A Cross-lingual Adversarial Dataset for Paraphrase Identification]]

===== Paraphrase Generation =====
  * [[https://huggingface.co/mrm8488/t5-small-finetuned-quora-for-paraphrasing|Google's T5 fine-tuned on 
Quora question pair dataset for Question Paraphrasing task]]

===== Miscellaneous Papers =====
  * [[https://arxiv.org/pdf/2210.11563.pdf|Tu et al 2022 - Dense Paraphrasing for Textual Enrichment]]

===== Question to Statement =====
Task: convert a question (and possibly the answer) into a statement.  Not quite paraphrase exactly, but related.

  * [[https://github.com/SunnyWay/question_to_statement]]
  * [[https://arxiv.org/pdf/1809.02922.pdf|Demszky et al 2018 - Transforming Question Answering Datasets Into Natural Language Inference Datasets]]. [[https://bit.ly/2OMm4vK|dataset]]

**Related pages:** [[Question Generation]]

===== Applications =====
  * [[https://aclanthology.org/P19-1610.pdf|Gan & Ng 2019 - Improving the Robustness of Question Answering Systems to Question Paraphrasing]]

===== Related Pages ======
  * [[Entailment]]
  * [[Semantics]]