Matches in SemOpenAlex for { <https://semopenalex.org/work/W3213373607> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W3213373607 abstract "Preserving the structural properties of trees or graphs when embedding them into a metric space allows for a high degree of interpretability, and has been shown beneficial for downstream tasks (e.g., hypernym detection, natural language inference, multimodal retrieval). However, whereas the majority of prior work looks at using structure-preserving embeddings when encoding a structure given as input, e.g., WordNet (Fellbaum, 1998), there is little exploration on how to use such embeddings when predicting one. We address this gap for two structure generation tasks, namely dependency and semantic parsing. We test the applicability of disk embeddings (Suzuki et al., 2019) that has been proposed for embedding Directed Acyclic Graphs (DAGs) but has not been tested on tasks that generate such structures. Our experimental results show that for both tasks the original disk embedding formulation leads to much worse performance when compared to non-structure-preserving baselines. We propose enhancements to this formulation and show that they almost close the performance gap for dependency parsing. However, the gap still remains notable for semantic parsing due to the complexity of meaning representation graphs, suggesting a challenge for generating interpretable semantic parse representations." @default.
- W3213373607 created "2021-11-22" @default.
- W3213373607 creator A5017305009 @default.
- W3213373607 creator A5027949576 @default.
- W3213373607 creator A5048077192 @default.
- W3213373607 creator A5084837026 @default.
- W3213373607 date "2021-01-01" @default.
- W3213373607 modified "2023-09-23" @default.
- W3213373607 title "An in-depth look at Euclidean disk embeddings for structure preserving parsing" @default.
- W3213373607 cites W2038721957 @default.
- W3213373607 cites W2153579005 @default.
- W3213373607 cites W2250375035 @default.
- W3213373607 cites W2250539671 @default.
- W3213373607 cites W2251803266 @default.
- W3213373607 cites W2493916176 @default.
- W3213373607 cites W2739827909 @default.
- W3213373607 cites W2741449676 @default.
- W3213373607 cites W2899024931 @default.
- W3213373607 cites W2910430412 @default.
- W3213373607 cites W2962936818 @default.
- W3213373607 cites W2963165302 @default.
- W3213373607 cites W2963690776 @default.
- W3213373607 cites W2963899908 @default.
- W3213373607 cites W2964109691 @default.
- W3213373607 cites W2970796523 @default.
- W3213373607 cites W2983040767 @default.
- W3213373607 cites W3018647120 @default.
- W3213373607 cites W3086282523 @default.
- W3213373607 cites W3103620756 @default.
- W3213373607 cites W3155549525 @default.
- W3213373607 cites W3167860749 @default.
- W3213373607 cites W3173655578 @default.
- W3213373607 cites W3173677700 @default.
- W3213373607 cites W3098739824 @default.
- W3213373607 cites W3196552872 @default.
- W3213373607 doi "https://doi.org/10.18653/v1/2021.blackboxnlp-1.21" @default.
- W3213373607 hasPublicationYear "2021" @default.
- W3213373607 type Work @default.
- W3213373607 sameAs 3213373607 @default.
- W3213373607 citedByCount "0" @default.
- W3213373607 crossrefType "proceedings-article" @default.
- W3213373607 hasAuthorship W3213373607A5017305009 @default.
- W3213373607 hasAuthorship W3213373607A5027949576 @default.
- W3213373607 hasAuthorship W3213373607A5048077192 @default.
- W3213373607 hasAuthorship W3213373607A5084837026 @default.
- W3213373607 hasBestOaLocation W32133736071 @default.
- W3213373607 hasConcept C154945302 @default.
- W3213373607 hasConcept C157659113 @default.
- W3213373607 hasConcept C164883195 @default.
- W3213373607 hasConcept C17744445 @default.
- W3213373607 hasConcept C186644900 @default.
- W3213373607 hasConcept C199539241 @default.
- W3213373607 hasConcept C204321447 @default.
- W3213373607 hasConcept C2776214188 @default.
- W3213373607 hasConcept C2776359362 @default.
- W3213373607 hasConcept C2781067378 @default.
- W3213373607 hasConcept C41008148 @default.
- W3213373607 hasConcept C41608201 @default.
- W3213373607 hasConcept C80444323 @default.
- W3213373607 hasConcept C94625758 @default.
- W3213373607 hasConceptScore W3213373607C154945302 @default.
- W3213373607 hasConceptScore W3213373607C157659113 @default.
- W3213373607 hasConceptScore W3213373607C164883195 @default.
- W3213373607 hasConceptScore W3213373607C17744445 @default.
- W3213373607 hasConceptScore W3213373607C186644900 @default.
- W3213373607 hasConceptScore W3213373607C199539241 @default.
- W3213373607 hasConceptScore W3213373607C204321447 @default.
- W3213373607 hasConceptScore W3213373607C2776214188 @default.
- W3213373607 hasConceptScore W3213373607C2776359362 @default.
- W3213373607 hasConceptScore W3213373607C2781067378 @default.
- W3213373607 hasConceptScore W3213373607C41008148 @default.
- W3213373607 hasConceptScore W3213373607C41608201 @default.
- W3213373607 hasConceptScore W3213373607C80444323 @default.
- W3213373607 hasConceptScore W3213373607C94625758 @default.
- W3213373607 hasLocation W32133736071 @default.
- W3213373607 hasOpenAccess W3213373607 @default.
- W3213373607 hasPrimaryLocation W32133736071 @default.
- W3213373607 hasRelatedWork W11273136 @default.
- W3213373607 hasRelatedWork W11436026 @default.
- W3213373607 hasRelatedWork W11862124 @default.
- W3213373607 hasRelatedWork W13286396 @default.
- W3213373607 hasRelatedWork W14306054 @default.
- W3213373607 hasRelatedWork W2141238 @default.
- W3213373607 hasRelatedWork W3205017 @default.
- W3213373607 hasRelatedWork W6270581 @default.
- W3213373607 hasRelatedWork W8097534 @default.
- W3213373607 hasRelatedWork W886613 @default.
- W3213373607 isParatext "false" @default.
- W3213373607 isRetracted "false" @default.
- W3213373607 magId "3213373607" @default.
- W3213373607 workType "article" @default.