Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385572912> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W4385572912 abstract "Since the meaning representations are detailed and accurate annotations which express fine-grained sequence-level semtantics, it is usually hard to train discriminative semantic parsers via Maximum Likelihood Estimation (MLE) in an autoregressive fashion. In this paper, we propose a semantic-aware contrastive learning algorithm, which can learn to distinguish fine-grained meaning representations and take the overall sequence-level semantic into consideration. Specifically, a multi-level online sampling algorithm is proposed to sample confusing and diverse instances. Three semantic-aware similarity functions are designed to accurately measure the distance between meaning representations as a whole. And a ranked contrastive loss is proposed to pull the representations of the semantic-identical instances together and push negative instances away. Experiments on two standard datasets show that our approach achieves significant improvements over MLE baselines and gets state-of-the-art performances by simply applying semantic-aware contrastive learning on a vanilla Seq2Seq model." @default.
- W4385572912 created "2023-08-05" @default.
- W4385572912 creator A5010347728 @default.
- W4385572912 creator A5018668652 @default.
- W4385572912 creator A5034236037 @default.
- W4385572912 creator A5067483347 @default.
- W4385572912 creator A5084951073 @default.
- W4385572912 date "2022-01-01" @default.
- W4385572912 modified "2023-10-14" @default.
- W4385572912 title "Semantic-aware Contrastive Learning for More Accurate Semantic Parsing" @default.
- W4385572912 doi "https://doi.org/10.18653/v1/2022.emnlp-main.269" @default.
- W4385572912 hasPublicationYear "2022" @default.
- W4385572912 type Work @default.
- W4385572912 citedByCount "0" @default.
- W4385572912 crossrefType "proceedings-article" @default.
- W4385572912 hasAuthorship W4385572912A5010347728 @default.
- W4385572912 hasAuthorship W4385572912A5018668652 @default.
- W4385572912 hasAuthorship W4385572912A5034236037 @default.
- W4385572912 hasAuthorship W4385572912A5067483347 @default.
- W4385572912 hasAuthorship W4385572912A5084951073 @default.
- W4385572912 hasBestOaLocation W43855729121 @default.
- W4385572912 hasConcept C103278499 @default.
- W4385572912 hasConcept C115961682 @default.
- W4385572912 hasConcept C130318100 @default.
- W4385572912 hasConcept C154945302 @default.
- W4385572912 hasConcept C15744967 @default.
- W4385572912 hasConcept C186644900 @default.
- W4385572912 hasConcept C204321447 @default.
- W4385572912 hasConcept C2777530160 @default.
- W4385572912 hasConcept C2780876879 @default.
- W4385572912 hasConcept C41008148 @default.
- W4385572912 hasConcept C542102704 @default.
- W4385572912 hasConcept C67277372 @default.
- W4385572912 hasConcept C97931131 @default.
- W4385572912 hasConceptScore W4385572912C103278499 @default.
- W4385572912 hasConceptScore W4385572912C115961682 @default.
- W4385572912 hasConceptScore W4385572912C130318100 @default.
- W4385572912 hasConceptScore W4385572912C154945302 @default.
- W4385572912 hasConceptScore W4385572912C15744967 @default.
- W4385572912 hasConceptScore W4385572912C186644900 @default.
- W4385572912 hasConceptScore W4385572912C204321447 @default.
- W4385572912 hasConceptScore W4385572912C2777530160 @default.
- W4385572912 hasConceptScore W4385572912C2780876879 @default.
- W4385572912 hasConceptScore W4385572912C41008148 @default.
- W4385572912 hasConceptScore W4385572912C542102704 @default.
- W4385572912 hasConceptScore W4385572912C67277372 @default.
- W4385572912 hasConceptScore W4385572912C97931131 @default.
- W4385572912 hasLocation W43855729121 @default.
- W4385572912 hasOpenAccess W4385572912 @default.
- W4385572912 hasPrimaryLocation W43855729121 @default.
- W4385572912 hasRelatedWork W1590308178 @default.
- W4385572912 hasRelatedWork W1971865446 @default.
- W4385572912 hasRelatedWork W2293457016 @default.
- W4385572912 hasRelatedWork W2366488229 @default.
- W4385572912 hasRelatedWork W2379525370 @default.
- W4385572912 hasRelatedWork W275552017 @default.
- W4385572912 hasRelatedWork W2798615242 @default.
- W4385572912 hasRelatedWork W2889841884 @default.
- W4385572912 hasRelatedWork W2977842567 @default.
- W4385572912 hasRelatedWork W576242789 @default.
- W4385572912 isParatext "false" @default.
- W4385572912 isRetracted "false" @default.
- W4385572912 workType "article" @default.