Matches in SemOpenAlex for { <https://semopenalex.org/work/W2987669390> ?p ?o ?g. }
- W2987669390 abstract "Non-extractive commonsense QA remains a challenging AI task, as it requires systems to reason about, synthesize, and gather disparate pieces of information, in order to generate responses to queries. Recent approaches on such tasks show increased performance, only when models are either pre-trained with additional information or when domain-specific heuristics are used, without any special consideration regarding the knowledge resource type. In this paper, we perform a survey of recent commonsense QA methods and we provide a systematic analysis of popular knowledge resources and knowledge-integration methods, across benchmarks from multiple commonsense datasets. Our results and analysis show that attention-based injection seems to be a preferable choice for knowledge integration and that the degree of domain overlap, between knowledge bases and datasets, plays a crucial role in determining model success." @default.
- W2987669390 created "2019-11-22" @default.
- W2987669390 creator A5036795131 @default.
- W2987669390 creator A5038055180 @default.
- W2987669390 creator A5045755805 @default.
- W2987669390 creator A5052185760 @default.
- W2987669390 creator A5063241234 @default.
- W2987669390 date "2019-01-01" @default.
- W2987669390 modified "2023-10-05" @default.
- W2987669390 title "Towards Generalizable Neuro-Symbolic Systems for Commonsense Question Answering" @default.
- W2987669390 cites W1492581928 @default.
- W2987669390 cites W1544827683 @default.
- W2987669390 cites W1850961690 @default.
- W2987669390 cites W1916313121 @default.
- W2987669390 cites W2107901333 @default.
- W2987669390 cites W2115792525 @default.
- W2987669390 cites W2161109368 @default.
- W2987669390 cites W2251883304 @default.
- W2987669390 cites W2551396370 @default.
- W2987669390 cites W2606964149 @default.
- W2987669390 cites W2740747242 @default.
- W2987669390 cites W2766508367 @default.
- W2987669390 cites W2788810909 @default.
- W2987669390 cites W2798858969 @default.
- W2987669390 cites W2806055002 @default.
- W2987669390 cites W2889317091 @default.
- W2987669390 cites W2892280852 @default.
- W2987669390 cites W2898662126 @default.
- W2987669390 cites W2912904516 @default.
- W2987669390 cites W2920665390 @default.
- W2987669390 cites W2946609015 @default.
- W2987669390 cites W2949276121 @default.
- W2987669390 cites W2950339735 @default.
- W2987669390 cites W2963115613 @default.
- W2987669390 cites W2963159690 @default.
- W2987669390 cites W2963341956 @default.
- W2987669390 cites W2963448850 @default.
- W2987669390 cites W2963748441 @default.
- W2987669390 cites W2963829073 @default.
- W2987669390 cites W2963871484 @default.
- W2987669390 cites W2963995027 @default.
- W2987669390 cites W2964207259 @default.
- W2987669390 cites W2965373594 @default.
- W2987669390 cites W2970597249 @default.
- W2987669390 cites W2970986510 @default.
- W2987669390 cites W2977745385 @default.
- W2987669390 cites W2983995706 @default.
- W2987669390 cites W2998374885 @default.
- W2987669390 cites W3035153870 @default.
- W2987669390 cites W3046423960 @default.
- W2987669390 doi "https://doi.org/10.18653/v1/d19-6003" @default.
- W2987669390 hasPublicationYear "2019" @default.
- W2987669390 type Work @default.
- W2987669390 sameAs 2987669390 @default.
- W2987669390 citedByCount "40" @default.
- W2987669390 countsByYear W29876693902020 @default.
- W2987669390 countsByYear W29876693902021 @default.
- W2987669390 countsByYear W29876693902022 @default.
- W2987669390 countsByYear W29876693902023 @default.
- W2987669390 crossrefType "proceedings-article" @default.
- W2987669390 hasAuthorship W2987669390A5036795131 @default.
- W2987669390 hasAuthorship W2987669390A5038055180 @default.
- W2987669390 hasAuthorship W2987669390A5045755805 @default.
- W2987669390 hasAuthorship W2987669390A5052185760 @default.
- W2987669390 hasAuthorship W2987669390A5063241234 @default.
- W2987669390 hasBestOaLocation W29876693901 @default.
- W2987669390 hasConcept C111919701 @default.
- W2987669390 hasConcept C119857082 @default.
- W2987669390 hasConcept C127705205 @default.
- W2987669390 hasConcept C134306372 @default.
- W2987669390 hasConcept C154945302 @default.
- W2987669390 hasConcept C162324750 @default.
- W2987669390 hasConcept C187736073 @default.
- W2987669390 hasConcept C193221554 @default.
- W2987669390 hasConcept C204321447 @default.
- W2987669390 hasConcept C206345919 @default.
- W2987669390 hasConcept C207685749 @default.
- W2987669390 hasConcept C23123220 @default.
- W2987669390 hasConcept C2522767166 @default.
- W2987669390 hasConcept C2780451532 @default.
- W2987669390 hasConcept C30542707 @default.
- W2987669390 hasConcept C31258907 @default.
- W2987669390 hasConcept C33923547 @default.
- W2987669390 hasConcept C36503486 @default.
- W2987669390 hasConcept C41008148 @default.
- W2987669390 hasConcept C44291984 @default.
- W2987669390 hasConceptScore W2987669390C111919701 @default.
- W2987669390 hasConceptScore W2987669390C119857082 @default.
- W2987669390 hasConceptScore W2987669390C127705205 @default.
- W2987669390 hasConceptScore W2987669390C134306372 @default.
- W2987669390 hasConceptScore W2987669390C154945302 @default.
- W2987669390 hasConceptScore W2987669390C162324750 @default.
- W2987669390 hasConceptScore W2987669390C187736073 @default.
- W2987669390 hasConceptScore W2987669390C193221554 @default.
- W2987669390 hasConceptScore W2987669390C204321447 @default.
- W2987669390 hasConceptScore W2987669390C206345919 @default.
- W2987669390 hasConceptScore W2987669390C207685749 @default.
- W2987669390 hasConceptScore W2987669390C23123220 @default.
- W2987669390 hasConceptScore W2987669390C2522767166 @default.
- W2987669390 hasConceptScore W2987669390C2780451532 @default.