Matches in SemOpenAlex for { <https://semopenalex.org/work/W3210060997> ?p ?o ?g. }
- W3210060997 abstract "State-of-the-art approaches to reasoning and question answering over knowledge graphs (KGs) usually scale with the number of edges and can only be applied effectively on small instance-dependent subgraphs. In this paper, we address this issue by showing that multi-hop and more complex logical reasoning can be accomplished separately without losing expressive power. Motivated by this insight, we propose an approach to multi-hop reasoning that scales linearly with the number of relation types in the graph, which is usually significantly smaller than the number of edges or nodes. This produces a set of candidate solutions that can be provably refined to recover the solution to the original problem. Our experiments on knowledge-based question answering show that our approach solves the multi-hop MetaQA dataset, achieves a new state-of-the-art on the more challenging WebQuestionsSP, is orders of magnitude more scalable than competitive approaches, and can achieve compositional generalization out of the training distribution." @default.
- W3210060997 created "2021-11-08" @default.
- W3210060997 creator A5018114555 @default.
- W3210060997 creator A5033465803 @default.
- W3210060997 creator A5044345184 @default.
- W3210060997 date "2021-10-27" @default.
- W3210060997 modified "2023-10-01" @default.
- W3210060997 title "SQALER: Scaling Question Answering by Decoupling Multi-Hop and Logical Reasoning" @default.
- W3210060997 cites W1426956448 @default.
- W3210060997 cites W1997315949 @default.
- W3210060997 cites W2064675550 @default.
- W3210060997 cites W2079785597 @default.
- W3210060997 cites W2093397547 @default.
- W3210060997 cites W2094728533 @default.
- W3210060997 cites W2127795553 @default.
- W3210060997 cites W2148721079 @default.
- W3210060997 cites W2171278097 @default.
- W3210060997 cites W2251079237 @default.
- W3210060997 cites W2738790068 @default.
- W3210060997 cites W2759136286 @default.
- W3210060997 cites W2769099080 @default.
- W3210060997 cites W2790224343 @default.
- W3210060997 cites W2866343820 @default.
- W3210060997 cites W2889583850 @default.
- W3210060997 cites W2890273554 @default.
- W3210060997 cites W2890961898 @default.
- W3210060997 cites W2891026113 @default.
- W3210060997 cites W2899813393 @default.
- W3210060997 cites W2908510526 @default.
- W3210060997 cites W2912521296 @default.
- W3210060997 cites W2922491632 @default.
- W3210060997 cites W2939515836 @default.
- W3210060997 cites W2945102109 @default.
- W3210060997 cites W2952355681 @default.
- W3210060997 cites W2960658350 @default.
- W3210060997 cites W2962886429 @default.
- W3210060997 cites W2963068946 @default.
- W3210060997 cites W2963096160 @default.
- W3210060997 cites W2963267799 @default.
- W3210060997 cites W2963341956 @default.
- W3210060997 cites W2963380480 @default.
- W3210060997 cites W2963403868 @default.
- W3210060997 cites W2963448850 @default.
- W3210060997 cites W2964152081 @default.
- W3210060997 cites W2964308564 @default.
- W3210060997 cites W2994695355 @default.
- W3210060997 cites W2995455605 @default.
- W3210060997 cites W2996094825 @default.
- W3210060997 cites W3005116477 @default.
- W3210060997 cites W3034862985 @default.
- W3210060997 cites W3037932933 @default.
- W3210060997 cites W3043172396 @default.
- W3210060997 cites W3082274269 @default.
- W3210060997 cites W3098406703 @default.
- W3210060997 cites W3101133462 @default.
- W3210060997 cites W3104150569 @default.
- W3210060997 cites W3152801999 @default.
- W3210060997 hasPublicationYear "2021" @default.
- W3210060997 type Work @default.
- W3210060997 sameAs 3210060997 @default.
- W3210060997 citedByCount "0" @default.
- W3210060997 crossrefType "posted-content" @default.
- W3210060997 hasAuthorship W3210060997A5018114555 @default.
- W3210060997 hasAuthorship W3210060997A5033465803 @default.
- W3210060997 hasAuthorship W3210060997A5044345184 @default.
- W3210060997 hasConcept C127413603 @default.
- W3210060997 hasConcept C132525143 @default.
- W3210060997 hasConcept C133731056 @default.
- W3210060997 hasConcept C154945302 @default.
- W3210060997 hasConcept C195818886 @default.
- W3210060997 hasConcept C205606062 @default.
- W3210060997 hasConcept C2524010 @default.
- W3210060997 hasConcept C25906391 @default.
- W3210060997 hasConcept C31258907 @default.
- W3210060997 hasConcept C33923547 @default.
- W3210060997 hasConcept C41008148 @default.
- W3210060997 hasConcept C44291984 @default.
- W3210060997 hasConcept C48044578 @default.
- W3210060997 hasConcept C77088390 @default.
- W3210060997 hasConcept C80444323 @default.
- W3210060997 hasConcept C99844830 @default.
- W3210060997 hasConceptScore W3210060997C127413603 @default.
- W3210060997 hasConceptScore W3210060997C132525143 @default.
- W3210060997 hasConceptScore W3210060997C133731056 @default.
- W3210060997 hasConceptScore W3210060997C154945302 @default.
- W3210060997 hasConceptScore W3210060997C195818886 @default.
- W3210060997 hasConceptScore W3210060997C205606062 @default.
- W3210060997 hasConceptScore W3210060997C2524010 @default.
- W3210060997 hasConceptScore W3210060997C25906391 @default.
- W3210060997 hasConceptScore W3210060997C31258907 @default.
- W3210060997 hasConceptScore W3210060997C33923547 @default.
- W3210060997 hasConceptScore W3210060997C41008148 @default.
- W3210060997 hasConceptScore W3210060997C44291984 @default.
- W3210060997 hasConceptScore W3210060997C48044578 @default.
- W3210060997 hasConceptScore W3210060997C77088390 @default.
- W3210060997 hasConceptScore W3210060997C80444323 @default.
- W3210060997 hasConceptScore W3210060997C99844830 @default.
- W3210060997 hasLocation W32100609971 @default.
- W3210060997 hasOpenAccess W3210060997 @default.
- W3210060997 hasPrimaryLocation W32100609971 @default.