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 61–70 of 209 papers

TitleStatusHype
Ask me in your own words: paraphrasing for multitask question answeringCode0
Learning Semantic Sentence Embeddings using Sequential Pair-wise DiscriminatorCode0
Continuous Decomposition of Granularity for Neural Paraphrase GenerationCode0
Generating Sentences from Disentangled Syntactic and Semantic SpacesCode0
Latent Paraphrasing: Perturbation on Layers Improves Knowledge Injection in Language ModelsCode0
Improving the Diversity of Unsupervised Paraphrasing with Embedding OutputsCode0
Improving Non-autoregressive Generation with Mixup TrainingCode0
'John ate 5 apples' != 'John ate some apples': Self-Supervised Paraphrase Quality Detection for Algebraic Word ProblemsCode0
Multilingual Lexical Simplification via Paraphrase GenerationCode0
PIP: Parse-Instructed Prefix for Syntactically Controlled Paraphrase GenerationCode0
Show:102550
← PrevPage 7 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