Matches in SemOpenAlex for { <https://semopenalex.org/work/W3091782007> ?p ?o ?g. }
- W3091782007 abstract "AMR-to-text generation is used to transduce Abstract Meaning Representation structures (AMR) into text. A key challenge in this task is to efficiently learn effective graph representations. Previously, Graph Convolution Networks (GCNs) were used to encode input AMRs, however, vanilla GCNs are not able to capture non-local information and additionally, they follow a local (first-order) information aggregation scheme. To account for these issues, larger and deeper GCN models are required to capture more complex interactions. In this paper, we introduce a dynamic fusion mechanism, proposing Lightweight Dynamic Graph Convolutional Networks (LDGCNs) that capture richer non-local interactions by synthesizing higher order information from the input graphs. We further develop two novel parameter saving strategies based on the group graph convolutions and weight tied convolutions to reduce memory usage and model complexity. With the help of these strategies, we are able to train a model with fewer parameters while maintaining the model capacity. Experiments demonstrate that LDGCNs outperform state-of-the-art models on two benchmark datasets for AMR-to-text generation with significantly fewer parameters." @default.
- W3091782007 created "2020-10-15" @default.
- W3091782007 creator A5012543243 @default.
- W3091782007 creator A5024343415 @default.
- W3091782007 creator A5030503109 @default.
- W3091782007 creator A5045807606 @default.
- W3091782007 creator A5061941569 @default.
- W3091782007 creator A5086664284 @default.
- W3091782007 creator A5086674741 @default.
- W3091782007 date "2020-10-09" @default.
- W3091782007 modified "2023-09-27" @default.
- W3091782007 title "Lightweight, Dynamic Graph Convolutional Networks for AMR-to-Text Generation" @default.
- W3091782007 cites W2064675550 @default.
- W3091782007 cites W2101105183 @default.
- W3091782007 cites W2133459682 @default.
- W3091782007 cites W2186615578 @default.
- W3091782007 cites W222053410 @default.
- W3091782007 cites W2252123671 @default.
- W3091782007 cites W2468355276 @default.
- W3091782007 cites W2549139847 @default.
- W3091782007 cites W2565245743 @default.
- W3091782007 cites W2612445135 @default.
- W3091782007 cites W2758950307 @default.
- W3091782007 cites W2778814079 @default.
- W3091782007 cites W2798749466 @default.
- W3091782007 cites W2866343820 @default.
- W3091782007 cites W2908336025 @default.
- W3091782007 cites W2922509574 @default.
- W3091782007 cites W2924961378 @default.
- W3091782007 cites W2932864487 @default.
- W3091782007 cites W2933565306 @default.
- W3091782007 cites W2945290078 @default.
- W3091782007 cites W2950898568 @default.
- W3091782007 cites W2951309718 @default.
- W3091782007 cites W2952032096 @default.
- W3091782007 cites W2952768212 @default.
- W3091782007 cites W2962784628 @default.
- W3091782007 cites W2962810718 @default.
- W3091782007 cites W2963016848 @default.
- W3091782007 cites W2963125010 @default.
- W3091782007 cites W2963403868 @default.
- W3091782007 cites W2963446712 @default.
- W3091782007 cites W2963662654 @default.
- W3091782007 cites W2963893572 @default.
- W3091782007 cites W2963970792 @default.
- W3091782007 cites W2964035651 @default.
- W3091782007 cites W2964114465 @default.
- W3091782007 cites W2964116568 @default.
- W3091782007 cites W2965469401 @default.
- W3091782007 cites W2970900903 @default.
- W3091782007 cites W2971087717 @default.
- W3091782007 cites W2971187756 @default.
- W3091782007 cites W2971323043 @default.
- W3091782007 cites W2971328361 @default.
- W3091782007 cites W2983995706 @default.
- W3091782007 cites W2996428491 @default.
- W3091782007 cites W2998702685 @default.
- W3091782007 cites W3003446182 @default.
- W3091782007 cites W3034497660 @default.
- W3091782007 cites W3105081824 @default.
- W3091782007 doi "https://doi.org/10.48550/arxiv.2010.04383" @default.
- W3091782007 hasPublicationYear "2020" @default.
- W3091782007 type Work @default.
- W3091782007 sameAs 3091782007 @default.
- W3091782007 citedByCount "0" @default.
- W3091782007 crossrefType "posted-content" @default.
- W3091782007 hasAuthorship W3091782007A5012543243 @default.
- W3091782007 hasAuthorship W3091782007A5024343415 @default.
- W3091782007 hasAuthorship W3091782007A5030503109 @default.
- W3091782007 hasAuthorship W3091782007A5045807606 @default.
- W3091782007 hasAuthorship W3091782007A5061941569 @default.
- W3091782007 hasAuthorship W3091782007A5086664284 @default.
- W3091782007 hasAuthorship W3091782007A5086674741 @default.
- W3091782007 hasBestOaLocation W30917820071 @default.
- W3091782007 hasConcept C104317684 @default.
- W3091782007 hasConcept C132525143 @default.
- W3091782007 hasConcept C13280743 @default.
- W3091782007 hasConcept C154945302 @default.
- W3091782007 hasConcept C17744445 @default.
- W3091782007 hasConcept C185592680 @default.
- W3091782007 hasConcept C185798385 @default.
- W3091782007 hasConcept C199539241 @default.
- W3091782007 hasConcept C205649164 @default.
- W3091782007 hasConcept C2776359362 @default.
- W3091782007 hasConcept C41008148 @default.
- W3091782007 hasConcept C45347329 @default.
- W3091782007 hasConcept C50644808 @default.
- W3091782007 hasConcept C55493867 @default.
- W3091782007 hasConcept C66746571 @default.
- W3091782007 hasConcept C80444323 @default.
- W3091782007 hasConcept C94625758 @default.
- W3091782007 hasConceptScore W3091782007C104317684 @default.
- W3091782007 hasConceptScore W3091782007C132525143 @default.
- W3091782007 hasConceptScore W3091782007C13280743 @default.
- W3091782007 hasConceptScore W3091782007C154945302 @default.
- W3091782007 hasConceptScore W3091782007C17744445 @default.
- W3091782007 hasConceptScore W3091782007C185592680 @default.
- W3091782007 hasConceptScore W3091782007C185798385 @default.
- W3091782007 hasConceptScore W3091782007C199539241 @default.
- W3091782007 hasConceptScore W3091782007C205649164 @default.