Matches in SemOpenAlex for { <https://semopenalex.org/work/W3129146692> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W3129146692 endingPage "106850" @default.
- W3129146692 startingPage "106850" @default.
- W3129146692 abstract "The purpose of unconditional text generation is to train a model with real sentences, then generate novel sentences of the same quality and diversity as the training data. However, when different metrics are used for comparing the methods of unconditional text generation, contradictory conclusions are drawn. The difficulty is that both the diversity and quality of the sample should be considered simultaneously when the models are evaluated. To solve this problem, a novel metric of distributional discrepancy (DD) is designed to evaluate generators based on the discrepancy between the generated and real training sentences. However, it cannot compute the DD directly because the distribution of real sentences is unavailable. Thus, we propose a method for estimating the DD by training a neural-network-based text classifier. For comparison, three existing metrics, bi-lingual evaluation understudy (BLEU) versus self-BLEU, language model score versus reverse language model score, and Fréchet embedding distance, along with the proposed DD, are used to evaluate two popular generative models of long short-term memory and generative pretrained transformer 2 on both synthetic and real data. Experimental results show that DD is significantly better than the three existing metrics for ranking these generative models." @default.
- W3129146692 created "2021-02-15" @default.
- W3129146692 creator A5024197883 @default.
- W3129146692 creator A5029032414 @default.
- W3129146692 creator A5043790437 @default.
- W3129146692 creator A5070559820 @default.
- W3129146692 creator A5079873685 @default.
- W3129146692 date "2021-04-01" @default.
- W3129146692 modified "2023-10-17" @default.
- W3129146692 title "Distributional discrepancy: A metric for unconditional text generation" @default.
- W3129146692 cites W1832693441 @default.
- W3129146692 cites W2064675550 @default.
- W3129146692 cites W2101105183 @default.
- W3129146692 cites W2118978333 @default.
- W3129146692 cites W2125814131 @default.
- W3129146692 cites W2265846598 @default.
- W3129146692 cites W2962843773 @default.
- W3129146692 cites W2963466651 @default.
- W3129146692 cites W2963730239 @default.
- W3129146692 cites W2974078243 @default.
- W3129146692 cites W2994200508 @default.
- W3129146692 cites W3017628311 @default.
- W3129146692 doi "https://doi.org/10.1016/j.knosys.2021.106850" @default.
- W3129146692 hasPublicationYear "2021" @default.
- W3129146692 type Work @default.
- W3129146692 sameAs 3129146692 @default.
- W3129146692 citedByCount "2" @default.
- W3129146692 countsByYear W31291466922022 @default.
- W3129146692 crossrefType "journal-article" @default.
- W3129146692 hasAuthorship W3129146692A5024197883 @default.
- W3129146692 hasAuthorship W3129146692A5029032414 @default.
- W3129146692 hasAuthorship W3129146692A5043790437 @default.
- W3129146692 hasAuthorship W3129146692A5070559820 @default.
- W3129146692 hasAuthorship W3129146692A5079873685 @default.
- W3129146692 hasBestOaLocation W31291466922 @default.
- W3129146692 hasConcept C119857082 @default.
- W3129146692 hasConcept C121332964 @default.
- W3129146692 hasConcept C137293760 @default.
- W3129146692 hasConcept C154945302 @default.
- W3129146692 hasConcept C162324750 @default.
- W3129146692 hasConcept C165801399 @default.
- W3129146692 hasConcept C167966045 @default.
- W3129146692 hasConcept C176217482 @default.
- W3129146692 hasConcept C204321447 @default.
- W3129146692 hasConcept C21547014 @default.
- W3129146692 hasConcept C2985684807 @default.
- W3129146692 hasConcept C39890363 @default.
- W3129146692 hasConcept C41008148 @default.
- W3129146692 hasConcept C50644808 @default.
- W3129146692 hasConcept C62520636 @default.
- W3129146692 hasConcept C66322947 @default.
- W3129146692 hasConcept C95623464 @default.
- W3129146692 hasConceptScore W3129146692C119857082 @default.
- W3129146692 hasConceptScore W3129146692C121332964 @default.
- W3129146692 hasConceptScore W3129146692C137293760 @default.
- W3129146692 hasConceptScore W3129146692C154945302 @default.
- W3129146692 hasConceptScore W3129146692C162324750 @default.
- W3129146692 hasConceptScore W3129146692C165801399 @default.
- W3129146692 hasConceptScore W3129146692C167966045 @default.
- W3129146692 hasConceptScore W3129146692C176217482 @default.
- W3129146692 hasConceptScore W3129146692C204321447 @default.
- W3129146692 hasConceptScore W3129146692C21547014 @default.
- W3129146692 hasConceptScore W3129146692C2985684807 @default.
- W3129146692 hasConceptScore W3129146692C39890363 @default.
- W3129146692 hasConceptScore W3129146692C41008148 @default.
- W3129146692 hasConceptScore W3129146692C50644808 @default.
- W3129146692 hasConceptScore W3129146692C62520636 @default.
- W3129146692 hasConceptScore W3129146692C66322947 @default.
- W3129146692 hasConceptScore W3129146692C95623464 @default.
- W3129146692 hasFunder F4320317325 @default.
- W3129146692 hasFunder F4320321001 @default.
- W3129146692 hasFunder F4320333335 @default.
- W3129146692 hasLocation W31291466921 @default.
- W3129146692 hasLocation W31291466922 @default.
- W3129146692 hasOpenAccess W3129146692 @default.
- W3129146692 hasPrimaryLocation W31291466921 @default.
- W3129146692 hasRelatedWork W1576360539 @default.
- W3129146692 hasRelatedWork W1652742547 @default.
- W3129146692 hasRelatedWork W2884815824 @default.
- W3129146692 hasRelatedWork W2939290240 @default.
- W3129146692 hasRelatedWork W3033942572 @default.
- W3129146692 hasRelatedWork W3093566930 @default.
- W3129146692 hasRelatedWork W3177920269 @default.
- W3129146692 hasRelatedWork W3212218375 @default.
- W3129146692 hasRelatedWork W4221055413 @default.
- W3129146692 hasRelatedWork W4310424780 @default.
- W3129146692 hasVolume "217" @default.
- W3129146692 isParatext "false" @default.
- W3129146692 isRetracted "false" @default.
- W3129146692 magId "3129146692" @default.
- W3129146692 workType "article" @default.