PPDB: The Paraphrase Database

By Juri Ganitkevitch, Benjamin Van Durme, Chris Callison-Burch. 2013

802 citations recorded by the source corpus.

Subfields: Natural Language Processing, Evaluation & Benchmarks

How researchers see this paper

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

In Natural Language Processing

In Evaluation & Benchmarks

Citations in the corpus

SemEval-2017 Task 1: Semantic Textual Similarity Multilingual and Crosslingual Focused Evaluation (2017, background): “Semantic inference tasks related to STS include textual entailment (Bentivogli et al., 2016; Bowman et al., 2015; Dagan et al., 2010), semantic relatedness (Bentivogli et al., 2016) and paraphrase detection (Xu et al., 2015; Ganitkevitch et al., 2013; Dolan et al., 2004).”
Supervised Learning of Universal Sentence Representations from Natural Language Inference Data (2017, methodology): “NLI as a supervised training set Our findings indicate that our model trained on SNLI obtains much better overall results than models trained on other supervised tasks such as COCO, dictionary definitions, NMT, PPDB (Ganitkevitch et al., 2013) and SST.”
Hypothesis Only Baselines in Natural Language Inference (2018, methodology): “FrameNet Plus (FN+) Using paraphrases from PPDB (Ganitkevitch et al., 2013 (1) a.”
Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks (2015, positive): “These systems are heavily feature engineered, generally using a combination of surface form overlap features and lexical distance features derived from WordNet or the Paraphrase Database (Ganitkevitch et al., 2013).”
BLEURT: Learning Robust Metrics for Text Generation (2020, methodology): “Backtranslation: We generate paraphrases and perturbations with backtranslation, that is, round trips from English to another language and then back to English with a translation model (Bannard and Callison-Burch, 2005; Ganitkevitch et al., 2013; Sennrich et al., 2016).”
Towards Universal Paraphrastic Sentence Embeddings (2015, methodology): “We consider six compositional architectures based on neural networks and train them on noisy phrase pairs from the Paraphrase Database (PPDB; Ganitkevitch et al., 2013 We consider models spanning the range of complexity from word averaging to LSTMs.”
Learning Natural Language Inference with LSTM (2015, background): “A future direction would be to incorporate other resources such as the paraphrase database (Ganitkevitch et al., 2013) into the learning process.”
Building a Semantic Parser Overnight (2015, background): “Our basic features mainly match words and bigrams in x and c, if they share a lemma or are aligned in the PPDB resource (Ganitkevitch et al., 2013).”
SemEval-2015 Task 2: Semantic Textual Similarity, English, Spanish and Pilot on Interpretability (2015, neutral): “, 2014) and PPDB (Ganitkevitch et al., 2013).”
Charagram: Embedding Words and Sentences via Character n-grams (2016, methodology): “We train on noisy paraphrase pairs from the Paraphrase Database (PPDB; Ganitkevitch et al., 2013) with anL2 regularized contrastive loss objective function, following the training procedure of Wieting et al. (2015) and Wieting et al.”
Ask the Right Questions: Active Question Reformulation with Reinforcement Learning (2017, methodology): “Extensions extract full sentences using a parsing based MT system [Li et al., 2009, Ganitkevitch et al., 2011, 2013].”
Learning Knowledge Graphs for Question Answering through Conversational Dialog (2015, methodology): “PPDB: Paraphrase relations from PPDB (Ganitkevitch et al., 2013) are derived by aligning bilingual parallel texts.”
Revisiting Recurrent Networks for Paraphrastic Sentence Embeddings (2017, positive): “We ﬁrst change data sources: rather than train on noisy phrase pairs from the Paraphrase Database (PPDB; Ganitkevitch et al., 2013), we use noisy sentence pairs obtained automatically by aligning Simple English to standard English Wikipedia (Coster and Kauchak, 2011).”
Good-Enough Compositional Data Augmentation (2019, background): “An alternative line of work on paraphrase-based data augmentation (Ganitkevitch et al., 2013; Iyyer et al., 2018) uses external, text-only resources to encourage robust interpretation of new inputs corresponding to known outputs.”
Discriminative Improvements to Distributional Sentence Similarity (2013, background): “These ideas may have applicability in other semantic similarity tasks, and we are also eager to apply them to new, large-scale automatically-induced paraphrase corpora (Ganitkevitch et al., 2013).”
Multi-Perspective Sentence Similarity Modeling with Convolutional Neural Networks (2015, methodology): “, 2015) only on the MSRP task since they were developed for paraphrase tasks, having been trained on word pairs from the Paraphrase Database (Ganitkevitch et al., 2013).”
Learning a Neural Semantic Parser from User Feedback (2017, methodology): “We perform ablation studies on the development sets (see Table 4) and ﬁnd that paraphrasing using PPDB consistently helps boost performance.”
Unsupervised Learning of Sentence Embeddings Using Compositional n-Gram Features (2017, methodology): “They generate sentence embeddings from diverse pre-trained word embeddings among which are unsupervised word embed-dings such as GloVe (Pennington et al., 2014) as well as supervised word embeddings such as paragram-SL999 (PSL) (Wieting et al., 2015) trained on the Paraphrase Database (Ganitkevitch ”