Matches in SemOpenAlex for { <https://semopenalex.org/work/W3037337776> ?p ?o ?g. }
- W3037337776 abstract "Large-scale language models pretrained on massive corpora of text, such as GPT-2, are powerful open-domain text generators. However, as our systematic examination reveals, it is still challenging for such models to generate coherent long passages of text ($>$1000 tokens), especially when the models are fine-tuned to the target domain on a small corpus. To overcome the limitation, we propose a simple but effective method of generating text in a progressive manner, inspired by generating images from low to high resolution. Our method first produces domain-specific content keywords and then progressively refines them into complete passages in multiple stages. The simple design allows our approach to take advantage of pretrained language models at each stage and effectively adapt to any target domain given only a small set of examples. We conduct a comprehensive empirical study with a broad set of evaluation metrics, and show that our approach significantly improves upon the fine-tuned GPT-2 in terms of domain-specific quality and sample efficiency. The coarse-to-fine nature of progressive generation also allows for a higher degree of control over the generated content." @default.
- W3037337776 created "2020-07-02" @default.
- W3037337776 creator A5000958452 @default.
- W3037337776 creator A5008835202 @default.
- W3037337776 creator A5009361486 @default.
- W3037337776 creator A5009547049 @default.
- W3037337776 creator A5085608858 @default.
- W3037337776 date "2020-06-28" @default.
- W3037337776 modified "2023-09-27" @default.
- W3037337776 title "Progressive Generation of Long Text." @default.
- W3037337776 cites W1956559956 @default.
- W3037337776 cites W1985610876 @default.
- W3037337776 cites W2132339004 @default.
- W3037337776 cites W2540413092 @default.
- W3037337776 cites W2594978815 @default.
- W3037337776 cites W2626778328 @default.
- W3037337776 cites W2757121784 @default.
- W3037337776 cites W2766527293 @default.
- W3037337776 cites W2787560479 @default.
- W3037337776 cites W2798664956 @default.
- W3037337776 cites W2798838035 @default.
- W3037337776 cites W2799054028 @default.
- W3037337776 cites W2915002151 @default.
- W3037337776 cites W2920538220 @default.
- W3037337776 cites W2925962955 @default.
- W3037337776 cites W2947898088 @default.
- W3037337776 cites W2948629866 @default.
- W3037337776 cites W2949555952 @default.
- W3037337776 cites W2949615363 @default.
- W3037337776 cites W2949888546 @default.
- W3037337776 cites W2951011353 @default.
- W3037337776 cites W2953141401 @default.
- W3037337776 cites W2953345635 @default.
- W3037337776 cites W2953740644 @default.
- W3037337776 cites W2963045354 @default.
- W3037337776 cites W2963248296 @default.
- W3037337776 cites W2963341956 @default.
- W3037337776 cites W2963592583 @default.
- W3037337776 cites W2963837967 @default.
- W3037337776 cites W2963878748 @default.
- W3037337776 cites W2963993699 @default.
- W3037337776 cites W2964110616 @default.
- W3037337776 cites W2964308564 @default.
- W3037337776 cites W2965282611 @default.
- W3037337776 cites W2967853777 @default.
- W3037337776 cites W2970102799 @default.
- W3037337776 cites W2982399380 @default.
- W3037337776 cites W2983962589 @default.
- W3037337776 cites W2985694911 @default.
- W3037337776 cites W2996287690 @default.
- W3037337776 cites W3033529678 @default.
- W3037337776 cites W3087983746 @default.
- W3037337776 cites W3088754460 @default.
- W3037337776 cites W3125056032 @default.
- W3037337776 hasPublicationYear "2020" @default.
- W3037337776 type Work @default.
- W3037337776 sameAs 3037337776 @default.
- W3037337776 citedByCount "19" @default.
- W3037337776 countsByYear W30373377762020 @default.
- W3037337776 countsByYear W30373377762021 @default.
- W3037337776 crossrefType "posted-content" @default.
- W3037337776 hasAuthorship W3037337776A5000958452 @default.
- W3037337776 hasAuthorship W3037337776A5008835202 @default.
- W3037337776 hasAuthorship W3037337776A5009361486 @default.
- W3037337776 hasAuthorship W3037337776A5009547049 @default.
- W3037337776 hasAuthorship W3037337776A5085608858 @default.
- W3037337776 hasConcept C111472728 @default.
- W3037337776 hasConcept C134306372 @default.
- W3037337776 hasConcept C137293760 @default.
- W3037337776 hasConcept C138885662 @default.
- W3037337776 hasConcept C154945302 @default.
- W3037337776 hasConcept C177264268 @default.
- W3037337776 hasConcept C199360897 @default.
- W3037337776 hasConcept C204321447 @default.
- W3037337776 hasConcept C2779530757 @default.
- W3037337776 hasConcept C2780586882 @default.
- W3037337776 hasConcept C33923547 @default.
- W3037337776 hasConcept C36503486 @default.
- W3037337776 hasConcept C41008148 @default.
- W3037337776 hasConceptScore W3037337776C111472728 @default.
- W3037337776 hasConceptScore W3037337776C134306372 @default.
- W3037337776 hasConceptScore W3037337776C137293760 @default.
- W3037337776 hasConceptScore W3037337776C138885662 @default.
- W3037337776 hasConceptScore W3037337776C154945302 @default.
- W3037337776 hasConceptScore W3037337776C177264268 @default.
- W3037337776 hasConceptScore W3037337776C199360897 @default.
- W3037337776 hasConceptScore W3037337776C204321447 @default.
- W3037337776 hasConceptScore W3037337776C2779530757 @default.
- W3037337776 hasConceptScore W3037337776C2780586882 @default.
- W3037337776 hasConceptScore W3037337776C33923547 @default.
- W3037337776 hasConceptScore W3037337776C36503486 @default.
- W3037337776 hasConceptScore W3037337776C41008148 @default.
- W3037337776 hasLocation W30373377761 @default.
- W3037337776 hasOpenAccess W3037337776 @default.
- W3037337776 hasPrimaryLocation W30373377761 @default.
- W3037337776 hasRelatedWork W2101105183 @default.
- W3037337776 hasRelatedWork W2888213795 @default.
- W3037337776 hasRelatedWork W2914855263 @default.
- W3037337776 hasRelatedWork W2962739339 @default.
- W3037337776 hasRelatedWork W2963096510 @default.