FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2985469912> ?p ?o ?g. }

• W2985469912 abstract "Adversarial training (AT) as a regularization method has proved its effectiveness in various tasks, such as image classification and text classification. Though there are successful applications of AT in many tasks of natural language processing (NLP), the mechanism behind it is still unclear. In this paper, we aim to apply AT on machine reading comprehension (MRC) and study its effects from multiple perspectives. We experiment with three different kinds of RC tasks: span-based RC, span-based RC with unanswerable questions and multi-choice RC. The experimental results show that the proposed method can improve the performance significantly and universally on SQuAD1.1, SQuAD2.0 and RACE. With virtual adversarial training (VAT), we explore the possibility of improving the RC models with semi-supervised learning and prove that examples from a different task are also beneficial. We also find that AT helps little in defending against artificial adversarial examples, but AT helps the model to learn better on examples that contain more low-frequency words." @default.
• W2985469912 created "2019-11-22" @default.
• W2985469912 creator A5019108029 @default.
• W2985469912 creator A5025839290 @default.
• W2985469912 creator A5051107002 @default.
• W2985469912 creator A5063089557 @default.
• W2985469912 creator A5069365665 @default.
• W2985469912 creator A5076447478 @default.
• W2985469912 date "2019-11-09" @default.
• W2985469912 modified "2023-09-27" @default.
• W2985469912 title "Improving Machine Reading Comprehension via Adversarial Training." @default.
• W2985469912 cites W1544827683 @default.
• W2985469912 cites W2551396370 @default.
• W2985469912 cites W2606964149 @default.
• W2985469912 cites W2760600531 @default.
• W2985469912 cites W2885478576 @default.
• W2985469912 cites W2891702339 @default.
• W2985469912 cites W2897076808 @default.
• W2985469912 cites W2920665390 @default.
• W2985469912 cites W2950651087 @default.
• W2985469912 cites W2951561177 @default.
• W2985469912 cites W2962808855 @default.
• W2985469912 cites W2963010846 @default.
• W2985469912 cites W2963083752 @default.
• W2985469912 cites W2963175042 @default.
• W2985469912 cites W2963207607 @default.
• W2985469912 cites W2963242170 @default.
• W2985469912 cites W2963323070 @default.
• W2985469912 cites W2963341956 @default.
• W2985469912 cites W2963403868 @default.
• W2985469912 cites W2963595025 @default.
• W2985469912 cites W2963748441 @default.
• W2985469912 cites W2963783970 @default.
• W2985469912 cites W2963938442 @default.
• W2985469912 cites W2963969878 @default.
• W2985469912 cites W2963997908 @default.
• W2985469912 cites W2964153729 @default.
• W2985469912 cites W2964185534 @default.
• W2985469912 cites W2964223283 @default.
• W2985469912 cites W2964267515 @default.
• W2985469912 cites W2965373594 @default.
• W2985469912 cites W2970597249 @default.
• W2985469912 cites W3011411500 @default.
• W2985469912 cites W3093419064 @default.
• W2985469912 hasPublicationYear "2019" @default.
• W2985469912 type Work @default.
• W2985469912 sameAs 2985469912 @default.
• W2985469912 citedByCount "9" @default.
• W2985469912 countsByYear W29854699122020 @default.
• W2985469912 countsByYear W29854699122021 @default.
• W2985469912 crossrefType "posted-content" @default.
• W2985469912 hasAuthorship W2985469912A5019108029 @default.
• W2985469912 hasAuthorship W2985469912A5025839290 @default.
• W2985469912 hasAuthorship W2985469912A5051107002 @default.
• W2985469912 hasAuthorship W2985469912A5063089557 @default.
• W2985469912 hasAuthorship W2985469912A5069365665 @default.
• W2985469912 hasAuthorship W2985469912A5076447478 @default.
• W2985469912 hasConcept C119857082 @default.
• W2985469912 hasConcept C127413603 @default.
• W2985469912 hasConcept C138885662 @default.
• W2985469912 hasConcept C154945302 @default.
• W2985469912 hasConcept C199360897 @default.
• W2985469912 hasConcept C201995342 @default.
• W2985469912 hasConcept C204321447 @default.
• W2985469912 hasConcept C2776135515 @default.
• W2985469912 hasConcept C2778780117 @default.
• W2985469912 hasConcept C2780451532 @default.
• W2985469912 hasConcept C37736160 @default.
• W2985469912 hasConcept C41008148 @default.
• W2985469912 hasConcept C41895202 @default.
• W2985469912 hasConcept C511192102 @default.
• W2985469912 hasConcept C554936623 @default.
• W2985469912 hasConceptScore W2985469912C119857082 @default.
• W2985469912 hasConceptScore W2985469912C127413603 @default.
• W2985469912 hasConceptScore W2985469912C138885662 @default.
• W2985469912 hasConceptScore W2985469912C154945302 @default.
• W2985469912 hasConceptScore W2985469912C199360897 @default.
• W2985469912 hasConceptScore W2985469912C201995342 @default.
• W2985469912 hasConceptScore W2985469912C204321447 @default.
• W2985469912 hasConceptScore W2985469912C2776135515 @default.
• W2985469912 hasConceptScore W2985469912C2778780117 @default.
• W2985469912 hasConceptScore W2985469912C2780451532 @default.
• W2985469912 hasConceptScore W2985469912C37736160 @default.
• W2985469912 hasConceptScore W2985469912C41008148 @default.
• W2985469912 hasConceptScore W2985469912C41895202 @default.
• W2985469912 hasConceptScore W2985469912C511192102 @default.
• W2985469912 hasConceptScore W2985469912C554936623 @default.
• W2985469912 hasLocation W29854699121 @default.
• W2985469912 hasOpenAccess W2985469912 @default.
• W2985469912 hasPrimaryLocation W29854699121 @default.
• W2985469912 hasRelatedWork W2250539671 @default.
• W2985469912 hasRelatedWork W2557764419 @default.
• W2985469912 hasRelatedWork W2899705998 @default.
• W2985469912 hasRelatedWork W2962808855 @default.
• W2985469912 hasRelatedWork W2963323070 @default.
• W2985469912 hasRelatedWork W2963339397 @default.
• W2985469912 hasRelatedWork W2963341956 @default.
• W2985469912 hasRelatedWork W2963597678 @default.
• W2985469912 hasRelatedWork W2963748441 @default.
• W2985469912 hasRelatedWork W2963783970 @default.

• next