Matches in SemOpenAlex for { <https://semopenalex.org/work/W4295281423> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4295281423 abstract "Named Entity Recognition (NER) in Few-Shot setting is imperative for entity tagging in low resource domains. Existing approaches only learn class-specific semantic features and intermediate representations from source domains. This affects generalizability to unseen target domains, resulting in suboptimal performances. To this end, we present CONTaiNER, a novel contrastive learning technique that optimizes the inter-token distribution distance for Few-Shot NER. Instead of optimizing class-specific attributes, CONTaiNER optimizes a generalized objective of differentiating between token categories based on their Gaussian-distributed embeddings. This effectively alleviates overfitting issues originating from training domains. Our experiments in several traditional test domains (OntoNotes, CoNLL'03, WNUT '17, GUM) and a new large scale Few-Shot NER dataset (Few-NERD) demonstrate that on average, CONTaiNER outperforms previous methods by 3%-13% absolute F1 points while showing consistent performance trends, even in challenging scenarios where previous approaches could not achieve appreciable performance." @default.
- W4295281423 created "2022-09-12" @default.
- W4295281423 creator A5039621382 @default.
- W4295281423 creator A5050352955 @default.
- W4295281423 creator A5065037360 @default.
- W4295281423 creator A5067607594 @default.
- W4295281423 date "2021-09-15" @default.
- W4295281423 modified "2023-10-16" @default.
- W4295281423 title "CONTaiNER: Few-Shot Named Entity Recognition via Contrastive Learning" @default.
- W4295281423 doi "https://doi.org/10.48550/arxiv.2109.07589" @default.
- W4295281423 hasPublicationYear "2021" @default.
- W4295281423 type Work @default.
- W4295281423 citedByCount "0" @default.
- W4295281423 crossrefType "posted-content" @default.
- W4295281423 hasAuthorship W4295281423A5039621382 @default.
- W4295281423 hasAuthorship W4295281423A5050352955 @default.
- W4295281423 hasAuthorship W4295281423A5065037360 @default.
- W4295281423 hasAuthorship W4295281423A5067607594 @default.
- W4295281423 hasBestOaLocation W42952814231 @default.
- W4295281423 hasConcept C105795698 @default.
- W4295281423 hasConcept C119857082 @default.
- W4295281423 hasConcept C127413603 @default.
- W4295281423 hasConcept C154945302 @default.
- W4295281423 hasConcept C162324750 @default.
- W4295281423 hasConcept C178790620 @default.
- W4295281423 hasConcept C185592680 @default.
- W4295281423 hasConcept C187736073 @default.
- W4295281423 hasConcept C204321447 @default.
- W4295281423 hasConcept C22019652 @default.
- W4295281423 hasConcept C27158222 @default.
- W4295281423 hasConcept C2777212361 @default.
- W4295281423 hasConcept C2778344882 @default.
- W4295281423 hasConcept C2779135771 @default.
- W4295281423 hasConcept C2780451532 @default.
- W4295281423 hasConcept C2781018962 @default.
- W4295281423 hasConcept C33923547 @default.
- W4295281423 hasConcept C38652104 @default.
- W4295281423 hasConcept C41008148 @default.
- W4295281423 hasConcept C4554734 @default.
- W4295281423 hasConcept C48145219 @default.
- W4295281423 hasConcept C50644808 @default.
- W4295281423 hasConcept C78519656 @default.
- W4295281423 hasConcept C96711827 @default.
- W4295281423 hasConceptScore W4295281423C105795698 @default.
- W4295281423 hasConceptScore W4295281423C119857082 @default.
- W4295281423 hasConceptScore W4295281423C127413603 @default.
- W4295281423 hasConceptScore W4295281423C154945302 @default.
- W4295281423 hasConceptScore W4295281423C162324750 @default.
- W4295281423 hasConceptScore W4295281423C178790620 @default.
- W4295281423 hasConceptScore W4295281423C185592680 @default.
- W4295281423 hasConceptScore W4295281423C187736073 @default.
- W4295281423 hasConceptScore W4295281423C204321447 @default.
- W4295281423 hasConceptScore W4295281423C22019652 @default.
- W4295281423 hasConceptScore W4295281423C27158222 @default.
- W4295281423 hasConceptScore W4295281423C2777212361 @default.
- W4295281423 hasConceptScore W4295281423C2778344882 @default.
- W4295281423 hasConceptScore W4295281423C2779135771 @default.
- W4295281423 hasConceptScore W4295281423C2780451532 @default.
- W4295281423 hasConceptScore W4295281423C2781018962 @default.
- W4295281423 hasConceptScore W4295281423C33923547 @default.
- W4295281423 hasConceptScore W4295281423C38652104 @default.
- W4295281423 hasConceptScore W4295281423C41008148 @default.
- W4295281423 hasConceptScore W4295281423C4554734 @default.
- W4295281423 hasConceptScore W4295281423C48145219 @default.
- W4295281423 hasConceptScore W4295281423C50644808 @default.
- W4295281423 hasConceptScore W4295281423C78519656 @default.
- W4295281423 hasConceptScore W4295281423C96711827 @default.
- W4295281423 hasLocation W42952814231 @default.
- W4295281423 hasOpenAccess W4295281423 @default.
- W4295281423 hasPrimaryLocation W42952814231 @default.
- W4295281423 hasRelatedWork W2405038964 @default.
- W4295281423 hasRelatedWork W2952732525 @default.
- W4295281423 hasRelatedWork W3012616397 @default.
- W4295281423 hasRelatedWork W3037880068 @default.
- W4295281423 hasRelatedWork W3107474891 @default.
- W4295281423 hasRelatedWork W4214765850 @default.
- W4295281423 hasRelatedWork W4287753411 @default.
- W4295281423 hasRelatedWork W4295281423 @default.
- W4295281423 hasRelatedWork W4310979480 @default.
- W4295281423 hasRelatedWork W4323240841 @default.
- W4295281423 isParatext "false" @default.
- W4295281423 isRetracted "false" @default.
- W4295281423 workType "article" @default.