Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385570466> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W4385570466 abstract "Attribute Value Extraction (AVE) aims to automatically obtain attribute value pairs from product descriptions to aid e-commerce. Despite the progressive performance of existing approaches in e-commerce platforms, they still suffer from two challenges: 1) difficulty in identifying values at different scales simultaneously; 2) easy confusion by some highly similar fine-grained attributes. This paper proposes a pre-training technique for AVE to address these issues. In particular, we first improve the conventional token-level masking strategy, guiding the language model to understand multi-scale values by recovering spans at the phrase and sentence level. Second, we apply clustering to build a challenging negative set for each example and design a pre-training objective based on contrastive learning to force the model to discriminate similar attributes. Comprehensive experiments show that our solution provides a significant improvement over traditional pre-trained models in the AVE task, and achieves state-of-the-art on four benchmarks." @default.
- W4385570466 created "2023-08-05" @default.
- W4385570466 creator A5000037580 @default.
- W4385570466 creator A5034606659 @default.
- W4385570466 creator A5036047645 @default.
- W4385570466 creator A5036220106 @default.
- W4385570466 creator A5059512471 @default.
- W4385570466 creator A5063230535 @default.
- W4385570466 creator A5064842058 @default.
- W4385570466 creator A5071278984 @default.
- W4385570466 creator A5071578712 @default.
- W4385570466 creator A5085015882 @default.
- W4385570466 date "2023-01-01" @default.
- W4385570466 modified "2023-09-24" @default.
- W4385570466 title "CoMave: Contrastive Pre-training with Multi-scale Masking for Attribute Value Extraction" @default.
- W4385570466 doi "https://doi.org/10.18653/v1/2023.findings-acl.373" @default.
- W4385570466 hasPublicationYear "2023" @default.
- W4385570466 type Work @default.
- W4385570466 citedByCount "0" @default.
- W4385570466 crossrefType "proceedings-article" @default.
- W4385570466 hasAuthorship W4385570466A5000037580 @default.
- W4385570466 hasAuthorship W4385570466A5034606659 @default.
- W4385570466 hasAuthorship W4385570466A5036047645 @default.
- W4385570466 hasAuthorship W4385570466A5036220106 @default.
- W4385570466 hasAuthorship W4385570466A5059512471 @default.
- W4385570466 hasAuthorship W4385570466A5063230535 @default.
- W4385570466 hasAuthorship W4385570466A5064842058 @default.
- W4385570466 hasAuthorship W4385570466A5071278984 @default.
- W4385570466 hasAuthorship W4385570466A5071578712 @default.
- W4385570466 hasAuthorship W4385570466A5085015882 @default.
- W4385570466 hasBestOaLocation W43855704661 @default.
- W4385570466 hasConcept C119857082 @default.
- W4385570466 hasConcept C121332964 @default.
- W4385570466 hasConcept C124101348 @default.
- W4385570466 hasConcept C127413603 @default.
- W4385570466 hasConcept C142362112 @default.
- W4385570466 hasConcept C153349607 @default.
- W4385570466 hasConcept C154945302 @default.
- W4385570466 hasConcept C177264268 @default.
- W4385570466 hasConcept C199360897 @default.
- W4385570466 hasConcept C201995342 @default.
- W4385570466 hasConcept C204321447 @default.
- W4385570466 hasConcept C2776224158 @default.
- W4385570466 hasConcept C2777402240 @default.
- W4385570466 hasConcept C2777530160 @default.
- W4385570466 hasConcept C2778755073 @default.
- W4385570466 hasConcept C2780451532 @default.
- W4385570466 hasConcept C28490314 @default.
- W4385570466 hasConcept C38652104 @default.
- W4385570466 hasConcept C41008148 @default.
- W4385570466 hasConcept C48145219 @default.
- W4385570466 hasConcept C62520636 @default.
- W4385570466 hasConcept C73555534 @default.
- W4385570466 hasConceptScore W4385570466C119857082 @default.
- W4385570466 hasConceptScore W4385570466C121332964 @default.
- W4385570466 hasConceptScore W4385570466C124101348 @default.
- W4385570466 hasConceptScore W4385570466C127413603 @default.
- W4385570466 hasConceptScore W4385570466C142362112 @default.
- W4385570466 hasConceptScore W4385570466C153349607 @default.
- W4385570466 hasConceptScore W4385570466C154945302 @default.
- W4385570466 hasConceptScore W4385570466C177264268 @default.
- W4385570466 hasConceptScore W4385570466C199360897 @default.
- W4385570466 hasConceptScore W4385570466C201995342 @default.
- W4385570466 hasConceptScore W4385570466C204321447 @default.
- W4385570466 hasConceptScore W4385570466C2776224158 @default.
- W4385570466 hasConceptScore W4385570466C2777402240 @default.
- W4385570466 hasConceptScore W4385570466C2777530160 @default.
- W4385570466 hasConceptScore W4385570466C2778755073 @default.
- W4385570466 hasConceptScore W4385570466C2780451532 @default.
- W4385570466 hasConceptScore W4385570466C28490314 @default.
- W4385570466 hasConceptScore W4385570466C38652104 @default.
- W4385570466 hasConceptScore W4385570466C41008148 @default.
- W4385570466 hasConceptScore W4385570466C48145219 @default.
- W4385570466 hasConceptScore W4385570466C62520636 @default.
- W4385570466 hasConceptScore W4385570466C73555534 @default.
- W4385570466 hasLocation W43855704661 @default.
- W4385570466 hasOpenAccess W4385570466 @default.
- W4385570466 hasPrimaryLocation W43855704661 @default.
- W4385570466 hasRelatedWork W1573537589 @default.
- W4385570466 hasRelatedWork W159132833 @default.
- W4385570466 hasRelatedWork W1597901428 @default.
- W4385570466 hasRelatedWork W2108705322 @default.
- W4385570466 hasRelatedWork W2251831371 @default.
- W4385570466 hasRelatedWork W2252152309 @default.
- W4385570466 hasRelatedWork W2369308426 @default.
- W4385570466 hasRelatedWork W2399585936 @default.
- W4385570466 hasRelatedWork W2400490375 @default.
- W4385570466 hasRelatedWork W2889818188 @default.
- W4385570466 isParatext "false" @default.
- W4385570466 isRetracted "false" @default.
- W4385570466 workType "article" @default.