FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2927431361> ?p ?o ?g. }

• W2927431361 abstract "Neural sequence-to-sequence models are currently the dominant approach in several natural language processing tasks, but require large parallel corpora. We present a sequence-to-sequence-to-sequence autoencoder (SEQˆ3), consisting of two chained encoder-decoder pairs, with words used as a sequence of discrete latent variables. We apply the proposed model to unsupervised abstractive sentence compression, where the first and last sequences are the input and reconstructed sentences, respectively, while the middle sequence is the compressed sentence. Constraining the length of the latent word sequences forces the model to distill important information from the input. A pretrained language model, acting as a prior over the latent sequences, encourages the compressed sentences to be human-readable. Continuous relaxations enable us to sample from categorical distributions, allowing gradient-based optimization, unlike alternatives that rely on reinforcement learning. The proposed model does not require parallel text-summary pairs, achieving promising results in unsupervised sentence compression on benchmark datasets." @default.
• W2927431361 created "2019-04-11" @default.
• W2927431361 creator A5030546839 @default.
• W2927431361 creator A5035665045 @default.
• W2927431361 creator A5069270736 @default.
• W2927431361 creator A5084949286 @default.
• W2927431361 date "2019-01-01" @default.
• W2927431361 modified "2023-09-26" @default.
• W2927431361 cites W1514535095 @default.
• W2927431361 cites W1902237438 @default.
• W2927431361 cites W1915251500 @default.
• W2927431361 cites W1999447745 @default.
• W2927431361 cites W2056609756 @default.
• W2927431361 cites W2064675550 @default.
• W2927431361 cites W2130942839 @default.
• W2927431361 cites W2154652894 @default.
• W2927431361 cites W2242818861 @default.
• W2927431361 cites W2250539671 @default.
• W2927431361 cites W23242996 @default.
• W2927431361 cites W2467173223 @default.
• W2927431361 cites W2526471240 @default.
• W2927431361 cites W2548228487 @default.
• W2927431361 cites W2583687993 @default.
• W2927431361 cites W2606974598 @default.
• W2927431361 cites W2609482285 @default.
• W2927431361 cites W2785910460 @default.
• W2927431361 cites W2885318751 @default.
• W2927431361 cites W2891836487 @default.
• W2927431361 cites W2899771611 @default.
• W2927431361 cites W2962701888 @default.
• W2927431361 cites W2962805889 @default.
• W2927431361 cites W2962824887 @default.
• W2927431361 cites W2962964385 @default.
• W2927431361 cites W2962965405 @default.
• W2927431361 cites W2963069010 @default.
• W2927431361 cites W2963104691 @default.
• W2927431361 cites W2963216553 @default.
• W2927431361 cites W2963223306 @default.
• W2927431361 cites W2963347649 @default.
• W2927431361 cites W2963506925 @default.
• W2927431361 cites W2963602293 @default.
• W2927431361 cites W2963681240 @default.
• W2927431361 cites W2963929190 @default.
• W2927431361 cites W2964121744 @default.
• W2927431361 cites W2964308564 @default.
• W2927431361 cites W3101913037 @default.
• W2927431361 doi "https://doi.org/10.18653/v1/n19-1071" @default.
• W2927431361 hasPublicationYear "2019" @default.
• W2927431361 type Work @default.
• W2927431361 sameAs 2927431361 @default.
• W2927431361 citedByCount "47" @default.
• W2927431361 countsByYear W29274313612019 @default.
• W2927431361 countsByYear W29274313612020 @default.
• W2927431361 countsByYear W29274313612021 @default.
• W2927431361 countsByYear W29274313612022 @default.
• W2927431361 crossrefType "proceedings-article" @default.
• W2927431361 hasAuthorship W2927431361A5030546839 @default.
• W2927431361 hasAuthorship W2927431361A5035665045 @default.
• W2927431361 hasAuthorship W2927431361A5069270736 @default.
• W2927431361 hasAuthorship W2927431361A5084949286 @default.
• W2927431361 hasConcept C101738243 @default.
• W2927431361 hasConcept C119857082 @default.
• W2927431361 hasConcept C13280743 @default.
• W2927431361 hasConcept C153180895 @default.
• W2927431361 hasConcept C154945302 @default.
• W2927431361 hasConcept C162324750 @default.
• W2927431361 hasConcept C185798385 @default.
• W2927431361 hasConcept C187736073 @default.
• W2927431361 hasConcept C204321447 @default.
• W2927431361 hasConcept C205649164 @default.
• W2927431361 hasConcept C2524010 @default.
• W2927431361 hasConcept C2777530160 @default.
• W2927431361 hasConcept C2778112365 @default.
• W2927431361 hasConcept C2780451532 @default.
• W2927431361 hasConcept C28490314 @default.
• W2927431361 hasConcept C33923547 @default.
• W2927431361 hasConcept C35639132 @default.
• W2927431361 hasConcept C40506919 @default.
• W2927431361 hasConcept C41008148 @default.
• W2927431361 hasConcept C50644808 @default.
• W2927431361 hasConcept C51167844 @default.
• W2927431361 hasConcept C5274069 @default.
• W2927431361 hasConcept C54355233 @default.
• W2927431361 hasConcept C78548338 @default.
• W2927431361 hasConcept C86803240 @default.
• W2927431361 hasConcept C90805587 @default.
• W2927431361 hasConceptScore W2927431361C101738243 @default.
• W2927431361 hasConceptScore W2927431361C119857082 @default.
• W2927431361 hasConceptScore W2927431361C13280743 @default.
• W2927431361 hasConceptScore W2927431361C153180895 @default.
• W2927431361 hasConceptScore W2927431361C154945302 @default.
• W2927431361 hasConceptScore W2927431361C162324750 @default.
• W2927431361 hasConceptScore W2927431361C185798385 @default.
• W2927431361 hasConceptScore W2927431361C187736073 @default.
• W2927431361 hasConceptScore W2927431361C204321447 @default.
• W2927431361 hasConceptScore W2927431361C205649164 @default.
• W2927431361 hasConceptScore W2927431361C2524010 @default.
• W2927431361 hasConceptScore W2927431361C2777530160 @default.
• W2927431361 hasConceptScore W2927431361C2778112365 @default.
• W2927431361 hasConceptScore W2927431361C2780451532 @default.

• next