Matches in SemOpenAlex for { <https://semopenalex.org/work/W3094328377> ?p ?o ?g. }
- W3094328377 abstract "Recently, knowledge graph (KG) augmented models have achieved noteworthy success on various commonsense reasoning tasks. However, KG edge (fact) sparsity and noisy edge extraction/generation often hinder models from obtaining useful knowledge to reason over. To address these issues, we propose a new KG-augmented model: Hybrid Graph Network (HGN). Unlike prior methods, HGN learns to jointly contextualize extracted and generated knowledge by reasoning over both within a unified graph structure. Given the task input context and an extracted KG subgraph, HGN is trained to generate embeddings for the subgraph's missing edges to form a hybrid graph, then reason over the hybrid graph while filtering out context-irrelevant edges. We demonstrate HGN's effectiveness through considerable performance gains across four commonsense reasoning benchmarks, plus a user study on edge validness and helpfulness." @default.
- W3094328377 created "2020-10-29" @default.
- W3094328377 creator A5006918187 @default.
- W3094328377 creator A5009408707 @default.
- W3094328377 creator A5009957887 @default.
- W3094328377 creator A5016933602 @default.
- W3094328377 creator A5019304605 @default.
- W3094328377 creator A5024078415 @default.
- W3094328377 creator A5049936379 @default.
- W3094328377 creator A5058478290 @default.
- W3094328377 creator A5066738831 @default.
- W3094328377 date "2020-10-24" @default.
- W3094328377 modified "2023-09-28" @default.
- W3094328377 title "Learning Contextualized Knowledge Structures for Commonsense Reasoning" @default.
- W3094328377 cites W1576350636 @default.
- W3094328377 cites W1975879668 @default.
- W3094328377 cites W2016089260 @default.
- W3094328377 cites W2101848544 @default.
- W3094328377 cites W2127795553 @default.
- W3094328377 cites W2251960799 @default.
- W3094328377 cites W2604314403 @default.
- W3094328377 cites W2781474777 @default.
- W3094328377 cites W2805387248 @default.
- W3094328377 cites W2805516822 @default.
- W3094328377 cites W2808877322 @default.
- W3094328377 cites W2890894339 @default.
- W3094328377 cites W2897037347 @default.
- W3094328377 cites W2948729509 @default.
- W3094328377 cites W2950339735 @default.
- W3094328377 cites W2962927471 @default.
- W3094328377 cites W2962965968 @default.
- W3094328377 cites W2963341956 @default.
- W3094328377 cites W2963464736 @default.
- W3094328377 cites W2963532541 @default.
- W3094328377 cites W2963858333 @default.
- W3094328377 cites W2963895422 @default.
- W3094328377 cites W2963907629 @default.
- W3094328377 cites W2963995027 @default.
- W3094328377 cites W2964152081 @default.
- W3094328377 cites W2964207259 @default.
- W3094328377 cites W2965373594 @default.
- W3094328377 cites W2970161131 @default.
- W3094328377 cites W2970476646 @default.
- W3094328377 cites W2971155257 @default.
- W3094328377 cites W2978170550 @default.
- W3094328377 cites W2983995706 @default.
- W3094328377 cites W2985671014 @default.
- W3094328377 cites W2986836624 @default.
- W3094328377 cites W2987669390 @default.
- W3094328377 cites W2994689640 @default.
- W3094328377 cites W2996428491 @default.
- W3094328377 cites W2996848635 @default.
- W3094328377 cites W2998231212 @default.
- W3094328377 cites W2998374885 @default.
- W3094328377 cites W2998631105 @default.
- W3094328377 cites W3015440086 @default.
- W3094328377 cites W3082274269 @default.
- W3094328377 cites W3088396740 @default.
- W3094328377 cites W3092049183 @default.
- W3094328377 cites W3092642435 @default.
- W3094328377 cites W3094815596 @default.
- W3094328377 cites W3097986428 @default.
- W3094328377 cites W3099655892 @default.
- W3094328377 cites W3101850416 @default.
- W3094328377 cites W3112571624 @default.
- W3094328377 cites W3134526511 @default.
- W3094328377 cites W3167128508 @default.
- W3094328377 cites W3172335055 @default.
- W3094328377 doi "https://doi.org/10.48550/arxiv.2010.12873" @default.
- W3094328377 hasPublicationYear "2020" @default.
- W3094328377 type Work @default.
- W3094328377 sameAs 3094328377 @default.
- W3094328377 citedByCount "4" @default.
- W3094328377 countsByYear W30943283772021 @default.
- W3094328377 crossrefType "posted-content" @default.
- W3094328377 hasAuthorship W3094328377A5006918187 @default.
- W3094328377 hasAuthorship W3094328377A5009408707 @default.
- W3094328377 hasAuthorship W3094328377A5009957887 @default.
- W3094328377 hasAuthorship W3094328377A5016933602 @default.
- W3094328377 hasAuthorship W3094328377A5019304605 @default.
- W3094328377 hasAuthorship W3094328377A5024078415 @default.
- W3094328377 hasAuthorship W3094328377A5049936379 @default.
- W3094328377 hasAuthorship W3094328377A5058478290 @default.
- W3094328377 hasAuthorship W3094328377A5066738831 @default.
- W3094328377 hasBestOaLocation W30943283771 @default.
- W3094328377 hasConcept C108010975 @default.
- W3094328377 hasConcept C119857082 @default.
- W3094328377 hasConcept C132525143 @default.
- W3094328377 hasConcept C151730666 @default.
- W3094328377 hasConcept C154945302 @default.
- W3094328377 hasConcept C15744967 @default.
- W3094328377 hasConcept C162307627 @default.
- W3094328377 hasConcept C193221554 @default.
- W3094328377 hasConcept C204321447 @default.
- W3094328377 hasConcept C2779343474 @default.
- W3094328377 hasConcept C2781265381 @default.
- W3094328377 hasConcept C2987255567 @default.
- W3094328377 hasConcept C30542707 @default.
- W3094328377 hasConcept C41008148 @default.
- W3094328377 hasConcept C4554734 @default.