Matches in SemOpenAlex for { <https://semopenalex.org/work/W3212758081> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W3212758081 abstract "Large-scale conversation models are turning to leveraging external knowledge to improve the factual accuracy in response generation. Considering the infeasibility to annotate the external knowledge for large-scale dialogue corpora, it is desirable to learn the knowledge selection and response generation in an unsupervised manner. In this paper, we propose PLATO-KAG (Knowledge-Augmented Generation), an unsupervised learning approach for end-to-end knowledge-grounded conversation modeling. For each dialogue context, the top-k relevant knowledge elements are selected and then employed in knowledge-grounded response generation. The two components of knowledge selection and response generation are optimized jointly and effectively under a balanced objective. Experimental results on two publicly available datasets validate the superiority of PLATO-KAG." @default.
- W3212758081 created "2021-11-22" @default.
- W3212758081 creator A5002705771 @default.
- W3212758081 creator A5003409838 @default.
- W3212758081 creator A5037949684 @default.
- W3212758081 creator A5038836690 @default.
- W3212758081 creator A5043069455 @default.
- W3212758081 creator A5062156254 @default.
- W3212758081 date "2021-01-01" @default.
- W3212758081 modified "2023-10-18" @default.
- W3212758081 title "PLATO-KAG: Unsupervised Knowledge-Grounded Conversation via Joint Modeling" @default.
- W3212758081 cites W1975879668 @default.
- W3212758081 cites W2127589108 @default.
- W3212758081 cites W2586847566 @default.
- W3212758081 cites W2891103209 @default.
- W3212758081 cites W2891826200 @default.
- W3212758081 cites W2915295540 @default.
- W3212758081 cites W2951508633 @default.
- W3212758081 cites W2963206148 @default.
- W3212758081 cites W2963475460 @default.
- W3212758081 cites W2963825865 @default.
- W3212758081 cites W2964121744 @default.
- W3212758081 cites W2964458951 @default.
- W3212758081 cites W2979478117 @default.
- W3212758081 cites W2995183464 @default.
- W3212758081 cites W2997300509 @default.
- W3212758081 cites W3005742798 @default.
- W3212758081 cites W3027879771 @default.
- W3212758081 cites W3034758256 @default.
- W3212758081 cites W3040352674 @default.
- W3212758081 cites W3099700870 @default.
- W3212758081 cites W3134891661 @default.
- W3212758081 cites W3155584966 @default.
- W3212758081 cites W3155807546 @default.
- W3212758081 cites W3186138538 @default.
- W3212758081 doi "https://doi.org/10.18653/v1/2021.nlp4convai-1.14" @default.
- W3212758081 hasPublicationYear "2021" @default.
- W3212758081 type Work @default.
- W3212758081 sameAs 3212758081 @default.
- W3212758081 citedByCount "1" @default.
- W3212758081 countsByYear W32127580812023 @default.
- W3212758081 crossrefType "proceedings-article" @default.
- W3212758081 hasAuthorship W3212758081A5002705771 @default.
- W3212758081 hasAuthorship W3212758081A5003409838 @default.
- W3212758081 hasAuthorship W3212758081A5037949684 @default.
- W3212758081 hasAuthorship W3212758081A5038836690 @default.
- W3212758081 hasAuthorship W3212758081A5043069455 @default.
- W3212758081 hasAuthorship W3212758081A5062156254 @default.
- W3212758081 hasBestOaLocation W32127580811 @default.
- W3212758081 hasConcept C151730666 @default.
- W3212758081 hasConcept C154945302 @default.
- W3212758081 hasConcept C15744967 @default.
- W3212758081 hasConcept C204321447 @default.
- W3212758081 hasConcept C2777200299 @default.
- W3212758081 hasConcept C2779343474 @default.
- W3212758081 hasConcept C41008148 @default.
- W3212758081 hasConcept C46312422 @default.
- W3212758081 hasConcept C81917197 @default.
- W3212758081 hasConcept C86803240 @default.
- W3212758081 hasConceptScore W3212758081C151730666 @default.
- W3212758081 hasConceptScore W3212758081C154945302 @default.
- W3212758081 hasConceptScore W3212758081C15744967 @default.
- W3212758081 hasConceptScore W3212758081C204321447 @default.
- W3212758081 hasConceptScore W3212758081C2777200299 @default.
- W3212758081 hasConceptScore W3212758081C2779343474 @default.
- W3212758081 hasConceptScore W3212758081C41008148 @default.
- W3212758081 hasConceptScore W3212758081C46312422 @default.
- W3212758081 hasConceptScore W3212758081C81917197 @default.
- W3212758081 hasConceptScore W3212758081C86803240 @default.
- W3212758081 hasLocation W32127580811 @default.
- W3212758081 hasOpenAccess W3212758081 @default.
- W3212758081 hasPrimaryLocation W32127580811 @default.
- W3212758081 hasRelatedWork W1552159754 @default.
- W3212758081 hasRelatedWork W2073045008 @default.
- W3212758081 hasRelatedWork W2274593003 @default.
- W3212758081 hasRelatedWork W2368651715 @default.
- W3212758081 hasRelatedWork W2496949096 @default.
- W3212758081 hasRelatedWork W2611614995 @default.
- W3212758081 hasRelatedWork W2789919619 @default.
- W3212758081 hasRelatedWork W3107474891 @default.
- W3212758081 hasRelatedWork W4367334022 @default.
- W3212758081 hasRelatedWork W1853743098 @default.
- W3212758081 isParatext "false" @default.
- W3212758081 isRetracted "false" @default.
- W3212758081 magId "3212758081" @default.
- W3212758081 workType "article" @default.