FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3046423960> ?p ?o ?g. }

• W3046423960 abstract "We focus on multiple-choice question answering (QA) tasks in subject areas such as science, where we require both broad background knowledge and the facts from the given subject-area reference corpus. In this work, we explore simple yet effective methods for exploiting two sources of external knowledge for subject-area QA. The first enriches the original subject-area reference corpus with relevant text snippets extracted from an open-domain resource (i.e., Wikipedia) that cover potentially ambiguous concepts in the question and answer options. As in other QA research, the second method simply increases the amount of training data by appending additional in-domain subject-area instances. Experiments on three challenging multiple-choice science QA tasks (i.e., ARC-Easy, ARC-Challenge, and OpenBookQA) demonstrate the effectiveness of our methods: in comparison to the previous state-of-the-art, we obtain absolute gains in accuracy of up to 8.1%, 13.0%, and 12.8%, respectively. While we observe consistent gains when we introduce knowledge from Wikipedia, we find that employing additional QA training instances is not uniformly helpful: performance degrades when the added instances exhibit a higher level of difficulty than the original training data. As one of the first studies on exploiting unstructured external knowledge for subject-area QA, we hope our methods, observations, and discussion of the exposed limitations may shed light on further developments in the area." @default.
• W3046423960 created "2020-08-07" @default.
• W3046423960 creator A5022703491 @default.
• W3046423960 creator A5034476404 @default.
• W3046423960 creator A5053739372 @default.
• W3046423960 creator A5063784915 @default.
• W3046423960 creator A5070511738 @default.
• W3046423960 creator A5075033889 @default.
• W3046423960 creator A5088339142 @default.
• W3046423960 date "2019-01-01" @default.
• W3046423960 modified "2023-09-23" @default.
• W3046423960 title "Improving Subject-Area Question Answering with External Knowledge" @default.
• W3046423960 cites W2007018602 @default.
• W3046423960 cites W2053468515 @default.
• W3046423960 cites W2123442489 @default.
• W3046423960 cites W2143017621 @default.
• W3046423960 cites W2252077675 @default.
• W3046423960 cites W2293174806 @default.
• W3046423960 cites W2460591548 @default.
• W3046423960 cites W2547185913 @default.
• W3046423960 cites W2606964149 @default.
• W3046423960 cites W2608309533 @default.
• W3046423960 cites W2609826708 @default.
• W3046423960 cites W2734823783 @default.
• W3046423960 cites W2741412196 @default.
• W3046423960 cites W2753329127 @default.
• W3046423960 cites W2773352682 @default.
• W3046423960 cites W2788496822 @default.
• W3046423960 cites W2788705621 @default.
• W3046423960 cites W2790767126 @default.
• W3046423960 cites W2794325560 @default.
• W3046423960 cites W2799042618 @default.
• W3046423960 cites W2799081691 @default.
• W3046423960 cites W2805387248 @default.
• W3046423960 cites W2889453388 @default.
• W3046423960 cites W2889729765 @default.
• W3046423960 cites W2890894339 @default.
• W3046423960 cites W2912924812 @default.
• W3046423960 cites W2945329331 @default.
• W3046423960 cites W2951534261 @default.
• W3046423960 cites W2951922096 @default.
• W3046423960 cites W2962829834 @default.
• W3046423960 cites W2962865973 @default.
• W3046423960 cites W2963123047 @default.
• W3046423960 cites W2963339397 @default.
• W3046423960 cites W2963341956 @default.
• W3046423960 cites W2963371565 @default.
• W3046423960 cites W2963547127 @default.
• W3046423960 cites W2963564796 @default.
• W3046423960 cites W2963748441 @default.
• W3046423960 cites W2963997623 @default.
• W3046423960 cites W2964222271 @default.
• W3046423960 cites W2977745385 @default.
• W3046423960 cites W2978990807 @default.
• W3046423960 doi "https://doi.org/10.18653/v1/d19-5804" @default.
• W3046423960 hasPublicationYear "2019" @default.
• W3046423960 type Work @default.
• W3046423960 sameAs 3046423960 @default.
• W3046423960 citedByCount "37" @default.
• W3046423960 countsByYear W30464239602019 @default.
• W3046423960 countsByYear W30464239602020 @default.
• W3046423960 countsByYear W30464239602021 @default.
• W3046423960 countsByYear W30464239602022 @default.
• W3046423960 countsByYear W30464239602023 @default.
• W3046423960 crossrefType "proceedings-article" @default.
• W3046423960 hasAuthorship W3046423960A5022703491 @default.
• W3046423960 hasAuthorship W3046423960A5034476404 @default.
• W3046423960 hasAuthorship W3046423960A5053739372 @default.
• W3046423960 hasAuthorship W3046423960A5063784915 @default.
• W3046423960 hasAuthorship W3046423960A5070511738 @default.
• W3046423960 hasAuthorship W3046423960A5075033889 @default.
• W3046423960 hasAuthorship W3046423960A5088339142 @default.
• W3046423960 hasBestOaLocation W30464239601 @default.
• W3046423960 hasConcept C105002631 @default.
• W3046423960 hasConcept C111472728 @default.
• W3046423960 hasConcept C120665830 @default.
• W3046423960 hasConcept C121332964 @default.
• W3046423960 hasConcept C127413603 @default.
• W3046423960 hasConcept C134306372 @default.
• W3046423960 hasConcept C136764020 @default.
• W3046423960 hasConcept C138885662 @default.
• W3046423960 hasConcept C154945302 @default.
• W3046423960 hasConcept C192209626 @default.
• W3046423960 hasConcept C206345919 @default.
• W3046423960 hasConcept C207685749 @default.
• W3046423960 hasConcept C23123220 @default.
• W3046423960 hasConcept C2522767166 @default.
• W3046423960 hasConcept C2777855551 @default.
• W3046423960 hasConcept C2780428219 @default.
• W3046423960 hasConcept C2780586882 @default.
• W3046423960 hasConcept C31258907 @default.
• W3046423960 hasConcept C33923547 @default.
• W3046423960 hasConcept C36503486 @default.
• W3046423960 hasConcept C41008148 @default.
• W3046423960 hasConcept C44291984 @default.
• W3046423960 hasConcept C58328972 @default.
• W3046423960 hasConcept C78519656 @default.
• W3046423960 hasConceptScore W3046423960C105002631 @default.
• W3046423960 hasConceptScore W3046423960C111472728 @default.
• W3046423960 hasConceptScore W3046423960C120665830 @default.

• next