Matches in SemOpenAlex for { <https://semopenalex.org/work/W2951873305> ?p ?o ?g. }
- W2951873305 abstract "A large number of reading comprehension (RC) datasets has been created recently, but little analysis has been done on whether they generalize to one another, and the extent to which existing datasets can be leveraged for improving performance on new ones. In this paper, we conduct such an investigation over ten RC datasets, training on one or more source RC datasets, and evaluating generalization, as well as transfer to a target RC dataset. We analyze the factors that contribute to generalization, and show that training on a source RC dataset and transferring to a target dataset substantially improves performance, even in the presence of powerful contextual representations from BERT (Devlin et al., 2019). We also find that training on multiple source RC datasets leads to robust generalization and transfer, and can reduce the cost of example collection for a new RC dataset. Following our analysis, we propose MultiQA, a BERT-based model, trained on multiple RC datasets, which leads to state-of-the-art performance on five RC datasets. We share our infrastructure for the benefit of the research community." @default.
- W2951873305 created "2019-06-27" @default.
- W2951873305 creator A5045872048 @default.
- W2951873305 creator A5070878072 @default.
- W2951873305 date "2019-01-01" @default.
- W2951873305 modified "2023-10-02" @default.
- W2951873305 title "MultiQA: An Empirical Investigation of Generalization and Transfer in Reading Comprehension" @default.
- W2951873305 cites W1544827683 @default.
- W2951873305 cites W2094728533 @default.
- W2951873305 cites W2125436846 @default.
- W2951873305 cites W2167482691 @default.
- W2951873305 cites W2250539671 @default.
- W2951873305 cites W2252016937 @default.
- W2951873305 cites W2557764419 @default.
- W2951873305 cites W2558203065 @default.
- W2951873305 cites W2572289264 @default.
- W2951873305 cites W2587528408 @default.
- W2951873305 cites W2606964149 @default.
- W2951873305 cites W2793978524 @default.
- W2951873305 cites W2799081691 @default.
- W2951873305 cites W2809324505 @default.
- W2951873305 cites W2884773089 @default.
- W2951873305 cites W2888302696 @default.
- W2951873305 cites W2889787757 @default.
- W2951873305 cites W2891850907 @default.
- W2951873305 cites W2892452601 @default.
- W2951873305 cites W2898700502 @default.
- W2951873305 cites W2916846408 @default.
- W2951873305 cites W2919420119 @default.
- W2951873305 cites W2927746189 @default.
- W2951873305 cites W2951815760 @default.
- W2951873305 cites W2962718483 @default.
- W2951873305 cites W2962739339 @default.
- W2951873305 cites W2962922117 @default.
- W2951873305 cites W2962925243 @default.
- W2951873305 cites W2963310665 @default.
- W2951873305 cites W2963339397 @default.
- W2951873305 cites W2963341956 @default.
- W2951873305 cites W2963350559 @default.
- W2951873305 cites W2963660055 @default.
- W2951873305 cites W2963748441 @default.
- W2951873305 cites W2963854351 @default.
- W2951873305 cites W2963866616 @default.
- W2951873305 cites W2963969878 @default.
- W2951873305 cites W2964222271 @default.
- W2951873305 cites W2964223283 @default.
- W2951873305 cites W3098057198 @default.
- W2951873305 doi "https://doi.org/10.18653/v1/p19-1485" @default.
- W2951873305 hasPublicationYear "2019" @default.
- W2951873305 type Work @default.
- W2951873305 sameAs 2951873305 @default.
- W2951873305 citedByCount "147" @default.
- W2951873305 countsByYear W29518733052019 @default.
- W2951873305 countsByYear W29518733052020 @default.
- W2951873305 countsByYear W29518733052021 @default.
- W2951873305 countsByYear W29518733052022 @default.
- W2951873305 countsByYear W29518733052023 @default.
- W2951873305 crossrefType "proceedings-article" @default.
- W2951873305 hasAuthorship W2951873305A5045872048 @default.
- W2951873305 hasAuthorship W2951873305A5070878072 @default.
- W2951873305 hasBestOaLocation W29518733051 @default.
- W2951873305 hasConcept C119857082 @default.
- W2951873305 hasConcept C124101348 @default.
- W2951873305 hasConcept C134306372 @default.
- W2951873305 hasConcept C150899416 @default.
- W2951873305 hasConcept C154945302 @default.
- W2951873305 hasConcept C173608175 @default.
- W2951873305 hasConcept C177148314 @default.
- W2951873305 hasConcept C17744445 @default.
- W2951873305 hasConcept C199360897 @default.
- W2951873305 hasConcept C199539241 @default.
- W2951873305 hasConcept C204321447 @default.
- W2951873305 hasConcept C2776175482 @default.
- W2951873305 hasConcept C2778780117 @default.
- W2951873305 hasConcept C33923547 @default.
- W2951873305 hasConcept C41008148 @default.
- W2951873305 hasConcept C511192102 @default.
- W2951873305 hasConcept C554936623 @default.
- W2951873305 hasConceptScore W2951873305C119857082 @default.
- W2951873305 hasConceptScore W2951873305C124101348 @default.
- W2951873305 hasConceptScore W2951873305C134306372 @default.
- W2951873305 hasConceptScore W2951873305C150899416 @default.
- W2951873305 hasConceptScore W2951873305C154945302 @default.
- W2951873305 hasConceptScore W2951873305C173608175 @default.
- W2951873305 hasConceptScore W2951873305C177148314 @default.
- W2951873305 hasConceptScore W2951873305C17744445 @default.
- W2951873305 hasConceptScore W2951873305C199360897 @default.
- W2951873305 hasConceptScore W2951873305C199539241 @default.
- W2951873305 hasConceptScore W2951873305C204321447 @default.
- W2951873305 hasConceptScore W2951873305C2776175482 @default.
- W2951873305 hasConceptScore W2951873305C2778780117 @default.
- W2951873305 hasConceptScore W2951873305C33923547 @default.
- W2951873305 hasConceptScore W2951873305C41008148 @default.
- W2951873305 hasConceptScore W2951873305C511192102 @default.
- W2951873305 hasConceptScore W2951873305C554936623 @default.
- W2951873305 hasLocation W29518733051 @default.
- W2951873305 hasLocation W29518733052 @default.
- W2951873305 hasOpenAccess W2951873305 @default.
- W2951873305 hasPrimaryLocation W29518733051 @default.
- W2951873305 hasRelatedWork W2097810762 @default.