Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386824976> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4386824976 endingPage "15" @default.
- W4386824976 startingPage "1" @default.
- W4386824976 abstract "Question answering requiring numerical reasoning, which generally involves symbolic operations such as sorting, counting, and addition, is a challenging task. To address such a problem, existing mixture-of-experts (MoE)-based methods design several specific answer predictors to handle different types of questions and achieve promising performance. However, they ignore the modeling and exploitation of fine-grained reasoning-related operations to support numerical reasoning, encountering the inadequacy in reasoning capability and interpretability. To alleviate this issue, we propose OPERA, an operation-augmented numerical reasoning framework. Concretely, we systematically define a scalable operation set to model numerical reasoning. We first identify reasoning-related operations based on context and then softly execute them to imitate the answer reasoning procedure via an operation-aware cross-attention mechanism. Finally, we utilize the operation-augmented semantic representation of execution results to support answer prediction. We verify the effectiveness and generalization of OPERA in two scenarios with different knowledge sources and reasoning capabilities. Specifically, we conduct extensive experiments on two textual datasets, DROP and RACENum, and a table-text hybrid dataset TAT-QA. Experiment results show that OPERA outperforms previous strong methods on the DROP, RACENum, and TAT-QA datasets. Further, we statistically and visually analyze its interpretability." @default.
- W4386824976 created "2023-09-19" @default.
- W4386824976 creator A5011555779 @default.
- W4386824976 creator A5013785156 @default.
- W4386824976 creator A5038560914 @default.
- W4386824976 creator A5047257329 @default.
- W4386824976 creator A5048159038 @default.
- W4386824976 date "2023-01-01" @default.
- W4386824976 modified "2023-09-27" @default.
- W4386824976 title "Operation-Augmented Numerical Reasoning for Question Answering" @default.
- W4386824976 doi "https://doi.org/10.1109/taslp.2023.3316448" @default.
- W4386824976 hasPublicationYear "2023" @default.
- W4386824976 type Work @default.
- W4386824976 citedByCount "0" @default.
- W4386824976 crossrefType "journal-article" @default.
- W4386824976 hasAuthorship W4386824976A5011555779 @default.
- W4386824976 hasAuthorship W4386824976A5013785156 @default.
- W4386824976 hasAuthorship W4386824976A5038560914 @default.
- W4386824976 hasAuthorship W4386824976A5047257329 @default.
- W4386824976 hasAuthorship W4386824976A5048159038 @default.
- W4386824976 hasConcept C119857082 @default.
- W4386824976 hasConcept C151730666 @default.
- W4386824976 hasConcept C154945302 @default.
- W4386824976 hasConcept C161301231 @default.
- W4386824976 hasConcept C177264268 @default.
- W4386824976 hasConcept C199360897 @default.
- W4386824976 hasConcept C20162079 @default.
- W4386824976 hasConcept C2779343474 @default.
- W4386824976 hasConcept C2781067378 @default.
- W4386824976 hasConcept C41008148 @default.
- W4386824976 hasConcept C44291984 @default.
- W4386824976 hasConcept C48044578 @default.
- W4386824976 hasConcept C77088390 @default.
- W4386824976 hasConcept C83725634 @default.
- W4386824976 hasConcept C86803240 @default.
- W4386824976 hasConcept C89288958 @default.
- W4386824976 hasConceptScore W4386824976C119857082 @default.
- W4386824976 hasConceptScore W4386824976C151730666 @default.
- W4386824976 hasConceptScore W4386824976C154945302 @default.
- W4386824976 hasConceptScore W4386824976C161301231 @default.
- W4386824976 hasConceptScore W4386824976C177264268 @default.
- W4386824976 hasConceptScore W4386824976C199360897 @default.
- W4386824976 hasConceptScore W4386824976C20162079 @default.
- W4386824976 hasConceptScore W4386824976C2779343474 @default.
- W4386824976 hasConceptScore W4386824976C2781067378 @default.
- W4386824976 hasConceptScore W4386824976C41008148 @default.
- W4386824976 hasConceptScore W4386824976C44291984 @default.
- W4386824976 hasConceptScore W4386824976C48044578 @default.
- W4386824976 hasConceptScore W4386824976C77088390 @default.
- W4386824976 hasConceptScore W4386824976C83725634 @default.
- W4386824976 hasConceptScore W4386824976C86803240 @default.
- W4386824976 hasConceptScore W4386824976C89288958 @default.
- W4386824976 hasFunder F4320321001 @default.
- W4386824976 hasLocation W43868249761 @default.
- W4386824976 hasOpenAccess W4386824976 @default.
- W4386824976 hasPrimaryLocation W43868249761 @default.
- W4386824976 hasRelatedWork W1486740420 @default.
- W4386824976 hasRelatedWork W1980812121 @default.
- W4386824976 hasRelatedWork W2099559366 @default.
- W4386824976 hasRelatedWork W2127482978 @default.
- W4386824976 hasRelatedWork W2750423720 @default.
- W4386824976 hasRelatedWork W4206534706 @default.
- W4386824976 hasRelatedWork W4229079080 @default.
- W4386824976 hasRelatedWork W4385957992 @default.
- W4386824976 hasRelatedWork W4385965371 @default.
- W4386824976 hasRelatedWork W4386025632 @default.
- W4386824976 isParatext "false" @default.
- W4386824976 isRetracted "false" @default.
- W4386824976 workType "article" @default.