Matches in SemOpenAlex for { <https://semopenalex.org/work/W2953142012> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W2953142012 abstract "In this paper, we introduce Adversarial-and-attention Network (A3Net) for Machine Reading Comprehension. This model extends existing approaches from two perspectives. First, adversarial training is applied to several target variables within the model, rather than only to the inputs or embeddings. We control the norm of adversarial perturbations according to the norm of original target variables, so that we can jointly add perturbations to several target variables during training. As an effective regularization method, adversarial training improves robustness and generalization of our model. Second, we propose a multi-layer attention network utilizing three kinds of high-efficiency attention mechanisms. Multi-layer attention conducts interaction between question and passage within each layer, which contributes to reasonable representation and understanding of the model. Combining these two contributions, we enhance the diversity of dataset and the information extracting ability of the model at the same time. Meanwhile, we construct A3Net for the WebQA dataset. Results show that our model outperforms the state-of-the-art models (improving Fuzzy Score from 73.50% to 77.0%)." @default.
- W2953142012 created "2019-06-27" @default.
- W2953142012 creator A5023599556 @default.
- W2953142012 creator A5057853630 @default.
- W2953142012 creator A5061724140 @default.
- W2953142012 creator A5068396083 @default.
- W2953142012 creator A5081895678 @default.
- W2953142012 creator A5090121996 @default.
- W2953142012 creator A5091025093 @default.
- W2953142012 date "2018-09-03" @default.
- W2953142012 modified "2023-09-23" @default.
- W2953142012 title "A3Net: Adversarial-and-Attention Network for Machine Reading Comprehension" @default.
- W2953142012 cites W2427527485 @default.
- W2953142012 cites W2495998536 @default.
- W2953142012 cites W2507756961 @default.
- W2953142012 cites W2516930406 @default.
- W2953142012 cites W2521709538 @default.
- W2953142012 cites W2625541525 @default.
- W2953142012 cites W2738015883 @default.
- W2953142012 cites W2760600531 @default.
- W2953142012 cites W2890177507 @default.
- W2953142012 cites W2949615363 @default.
- W2953142012 cites W2950748123 @default.
- W2953142012 cites W2963207607 @default.
- W2953142012 cites W2963769536 @default.
- W2953142012 cites W2964153729 @default.
- W2953142012 cites W2964298608 @default.
- W2953142012 cites W2964308564 @default.
- W2953142012 doi "https://doi.org/10.48550/arxiv.1809.00676" @default.
- W2953142012 hasPublicationYear "2018" @default.
- W2953142012 type Work @default.
- W2953142012 sameAs 2953142012 @default.
- W2953142012 citedByCount "0" @default.
- W2953142012 crossrefType "posted-content" @default.
- W2953142012 hasAuthorship W2953142012A5023599556 @default.
- W2953142012 hasAuthorship W2953142012A5057853630 @default.
- W2953142012 hasAuthorship W2953142012A5061724140 @default.
- W2953142012 hasAuthorship W2953142012A5068396083 @default.
- W2953142012 hasAuthorship W2953142012A5081895678 @default.
- W2953142012 hasAuthorship W2953142012A5090121996 @default.
- W2953142012 hasAuthorship W2953142012A5091025093 @default.
- W2953142012 hasBestOaLocation W29531420121 @default.
- W2953142012 hasConcept C104317684 @default.
- W2953142012 hasConcept C119857082 @default.
- W2953142012 hasConcept C154945302 @default.
- W2953142012 hasConcept C17744445 @default.
- W2953142012 hasConcept C185592680 @default.
- W2953142012 hasConcept C191795146 @default.
- W2953142012 hasConcept C199539241 @default.
- W2953142012 hasConcept C2776135515 @default.
- W2953142012 hasConcept C37736160 @default.
- W2953142012 hasConcept C41008148 @default.
- W2953142012 hasConcept C55493867 @default.
- W2953142012 hasConcept C63479239 @default.
- W2953142012 hasConceptScore W2953142012C104317684 @default.
- W2953142012 hasConceptScore W2953142012C119857082 @default.
- W2953142012 hasConceptScore W2953142012C154945302 @default.
- W2953142012 hasConceptScore W2953142012C17744445 @default.
- W2953142012 hasConceptScore W2953142012C185592680 @default.
- W2953142012 hasConceptScore W2953142012C191795146 @default.
- W2953142012 hasConceptScore W2953142012C199539241 @default.
- W2953142012 hasConceptScore W2953142012C2776135515 @default.
- W2953142012 hasConceptScore W2953142012C37736160 @default.
- W2953142012 hasConceptScore W2953142012C41008148 @default.
- W2953142012 hasConceptScore W2953142012C55493867 @default.
- W2953142012 hasConceptScore W2953142012C63479239 @default.
- W2953142012 hasLocation W29531420121 @default.
- W2953142012 hasOpenAccess W2953142012 @default.
- W2953142012 hasPrimaryLocation W29531420121 @default.
- W2953142012 hasRelatedWork W2610321374 @default.
- W2953142012 hasRelatedWork W2796902985 @default.
- W2953142012 hasRelatedWork W2946768379 @default.
- W2953142012 hasRelatedWork W3035729345 @default.
- W2953142012 hasRelatedWork W3193857078 @default.
- W2953142012 hasRelatedWork W4287549955 @default.
- W2953142012 hasRelatedWork W4303857474 @default.
- W2953142012 hasRelatedWork W4306177400 @default.
- W2953142012 hasRelatedWork W4311734044 @default.
- W2953142012 hasRelatedWork W4312741812 @default.
- W2953142012 isParatext "false" @default.
- W2953142012 isRetracted "false" @default.
- W2953142012 magId "2953142012" @default.
- W2953142012 workType "article" @default.