SOTAVerified

Paraphrase Generation

Paraphrase Generation involves transforming a natural language sentence to a new sentence, that has the same semantic meaning but a different syntactic or lexical surface form.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 21–30 of 209 papers

TitleStatusHype
Contrastive Representation Learning for Exemplar-Guided Paraphrase GenerationCode1
Neural semi-Markov CRF for Monolingual Word AlignmentCode1
Factorising Meaning and Form for Intent-Preserving ParaphrasingCode1
Generating Syntactically Controlled Paraphrases without Using Annotated Parallel PairsCode1
ParaSCI: A Large Scientific Paraphrase Dataset for Longer Paraphrase GenerationCode1
Latent Template Induction with Gumbel-CRFsCode1
Reformulating Unsupervised Style Transfer as Paraphrase GenerationCode1
iNLTK: Natural Language Toolkit for Indic LanguagesCode1
Paraphrase Generation as Zero-Shot Multilingual Translation: Disentangling Semantic Similarity from Lexical and Syntactic DiversityCode1
Simultaneous Translation and Paraphrase for Language EducationCode1
Show:102550
← PrevPage 3 of 21Next →
All datasetsParalexQuora Question PairsMSCOCO

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1HRQ-VAEiBLEU24.93—Unverified
2SeparatoriBLEU14.84—Unverified
#ModelMetricClaimedVerifiedStatus
1HRQ-VAEiBLEU18.42—Unverified
2SeparatoriBLEU5.84—Unverified
#ModelMetricClaimedVerifiedStatus
1HRQ-VAEBLEU27.9—Unverified