Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366460890> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W4366460890 abstract "Commonsense question-answering (QA) methods combine the power of pre-trained Language Models (LM) with the reasoning provided by Knowledge Graphs (KG). A typical approach collects nodes relevant to the QA pair from a KG to form a Working Graph (WG) followed by reasoning using Graph Neural Networks(GNNs). This faces two major challenges: (i) it is difficult to capture all the information from the QA in the WG, and (ii) the WG contains some irrelevant nodes from the KG. To address these, we propose GrapeQA with two simple improvements on the WG: (i) Prominent Entities for Graph Augmentation identifies relevant text chunks from the QA pair and augments the WG with corresponding latent representations from the LM, and (ii) Context-Aware Node Pruning removes nodes that are less relevant to the QA pair. We evaluate our results on OpenBookQA, CommonsenseQA and MedQA-USMLE and see that GrapeQA shows consistent improvements over its LM + KG predecessor (QA-GNN in particular) and large improvements on OpenBookQA." @default.
- W4366460890 created "2023-04-22" @default.
- W4366460890 creator A5003125633 @default.
- W4366460890 creator A5051444929 @default.
- W4366460890 creator A5052901001 @default.
- W4366460890 creator A5053109550 @default.
- W4366460890 creator A5063029157 @default.
- W4366460890 creator A5073786634 @default.
- W4366460890 date "2023-03-22" @default.
- W4366460890 modified "2023-09-26" @default.
- W4366460890 title "GrapeQA: GRaph Augmentation and Pruning to Enhance Question-Answering" @default.
- W4366460890 doi "https://doi.org/10.48550/arxiv.2303.12320" @default.
- W4366460890 hasPublicationYear "2023" @default.
- W4366460890 type Work @default.
- W4366460890 citedByCount "0" @default.
- W4366460890 crossrefType "posted-content" @default.
- W4366460890 hasAuthorship W4366460890A5003125633 @default.
- W4366460890 hasAuthorship W4366460890A5051444929 @default.
- W4366460890 hasAuthorship W4366460890A5052901001 @default.
- W4366460890 hasAuthorship W4366460890A5053109550 @default.
- W4366460890 hasAuthorship W4366460890A5063029157 @default.
- W4366460890 hasAuthorship W4366460890A5073786634 @default.
- W4366460890 hasBestOaLocation W43664608901 @default.
- W4366460890 hasConcept C108010975 @default.
- W4366460890 hasConcept C119857082 @default.
- W4366460890 hasConcept C132525143 @default.
- W4366460890 hasConcept C137293760 @default.
- W4366460890 hasConcept C151730666 @default.
- W4366460890 hasConcept C154945302 @default.
- W4366460890 hasConcept C204321447 @default.
- W4366460890 hasConcept C2779343474 @default.
- W4366460890 hasConcept C2987255567 @default.
- W4366460890 hasConcept C41008148 @default.
- W4366460890 hasConcept C44291984 @default.
- W4366460890 hasConcept C6557445 @default.
- W4366460890 hasConcept C80444323 @default.
- W4366460890 hasConcept C86803240 @default.
- W4366460890 hasConceptScore W4366460890C108010975 @default.
- W4366460890 hasConceptScore W4366460890C119857082 @default.
- W4366460890 hasConceptScore W4366460890C132525143 @default.
- W4366460890 hasConceptScore W4366460890C137293760 @default.
- W4366460890 hasConceptScore W4366460890C151730666 @default.
- W4366460890 hasConceptScore W4366460890C154945302 @default.
- W4366460890 hasConceptScore W4366460890C204321447 @default.
- W4366460890 hasConceptScore W4366460890C2779343474 @default.
- W4366460890 hasConceptScore W4366460890C2987255567 @default.
- W4366460890 hasConceptScore W4366460890C41008148 @default.
- W4366460890 hasConceptScore W4366460890C44291984 @default.
- W4366460890 hasConceptScore W4366460890C6557445 @default.
- W4366460890 hasConceptScore W4366460890C80444323 @default.
- W4366460890 hasConceptScore W4366460890C86803240 @default.
- W4366460890 hasLocation W43664608901 @default.
- W4366460890 hasOpenAccess W4366460890 @default.
- W4366460890 hasPrimaryLocation W43664608901 @default.
- W4366460890 hasRelatedWork W128392744 @default.
- W4366460890 hasRelatedWork W207304934 @default.
- W4366460890 hasRelatedWork W2747680751 @default.
- W4366460890 hasRelatedWork W3107474891 @default.
- W4366460890 hasRelatedWork W3137014599 @default.
- W4366460890 hasRelatedWork W4283807154 @default.
- W4366460890 hasRelatedWork W4287634190 @default.
- W4366460890 hasRelatedWork W4297899248 @default.
- W4366460890 hasRelatedWork W4360831820 @default.
- W4366460890 hasRelatedWork W4366460890 @default.
- W4366460890 isParatext "false" @default.
- W4366460890 isRetracted "false" @default.
- W4366460890 workType "article" @default.