Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285298351> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W4285298351 abstract "In order to equip NLP systems with ‘selective prediction’ capability, several task-specific approaches have been proposed. However, which approaches work best across tasks or even if they consistently outperform the simplest baseline MaxProb remains to be explored. To this end, we systematically study selective prediction in a large-scale setup of 17 datasets across several NLP tasks. Through comprehensive experiments under in-domain (IID), out-of-domain (OOD), and adversarial (ADV) settings, we show that despite leveraging additional resources (held-out data/computation), none of the existing approaches consistently and considerably outperforms MaxProb in all three settings. Furthermore, their performance does not translate well across tasks. For instance, Monte-Carlo Dropout outperforms all other approaches on Duplicate Detection datasets but does not fare well on NLI datasets, especially in the OOD setting. Thus, we recommend that future selective prediction approaches should be evaluated across tasks and settings for reliable estimation of their capabilities." @default.
- W4285298351 created "2022-07-14" @default.
- W4285298351 creator A5004217423 @default.
- W4285298351 creator A5063722751 @default.
- W4285298351 creator A5083735830 @default.
- W4285298351 date "2022-01-01" @default.
- W4285298351 modified "2023-10-14" @default.
- W4285298351 title "Investigating Selective Prediction Approaches Across Several Tasks in IID, OOD, and Adversarial Settings" @default.
- W4285298351 doi "https://doi.org/10.18653/v1/2022.findings-acl.158" @default.
- W4285298351 hasPublicationYear "2022" @default.
- W4285298351 type Work @default.
- W4285298351 citedByCount "5" @default.
- W4285298351 countsByYear W42852983512022 @default.
- W4285298351 countsByYear W42852983512023 @default.
- W4285298351 crossrefType "proceedings-article" @default.
- W4285298351 hasAuthorship W4285298351A5004217423 @default.
- W4285298351 hasAuthorship W4285298351A5063722751 @default.
- W4285298351 hasAuthorship W4285298351A5083735830 @default.
- W4285298351 hasBestOaLocation W42852983511 @default.
- W4285298351 hasConcept C111368507 @default.
- W4285298351 hasConcept C11413529 @default.
- W4285298351 hasConcept C119857082 @default.
- W4285298351 hasConcept C124101348 @default.
- W4285298351 hasConcept C12725497 @default.
- W4285298351 hasConcept C127313418 @default.
- W4285298351 hasConcept C134306372 @default.
- W4285298351 hasConcept C154945302 @default.
- W4285298351 hasConcept C162324750 @default.
- W4285298351 hasConcept C187736073 @default.
- W4285298351 hasConcept C2776145597 @default.
- W4285298351 hasConcept C2780451532 @default.
- W4285298351 hasConcept C33923547 @default.
- W4285298351 hasConcept C36503486 @default.
- W4285298351 hasConcept C37736160 @default.
- W4285298351 hasConcept C41008148 @default.
- W4285298351 hasConcept C45374587 @default.
- W4285298351 hasConceptScore W4285298351C111368507 @default.
- W4285298351 hasConceptScore W4285298351C11413529 @default.
- W4285298351 hasConceptScore W4285298351C119857082 @default.
- W4285298351 hasConceptScore W4285298351C124101348 @default.
- W4285298351 hasConceptScore W4285298351C12725497 @default.
- W4285298351 hasConceptScore W4285298351C127313418 @default.
- W4285298351 hasConceptScore W4285298351C134306372 @default.
- W4285298351 hasConceptScore W4285298351C154945302 @default.
- W4285298351 hasConceptScore W4285298351C162324750 @default.
- W4285298351 hasConceptScore W4285298351C187736073 @default.
- W4285298351 hasConceptScore W4285298351C2776145597 @default.
- W4285298351 hasConceptScore W4285298351C2780451532 @default.
- W4285298351 hasConceptScore W4285298351C33923547 @default.
- W4285298351 hasConceptScore W4285298351C36503486 @default.
- W4285298351 hasConceptScore W4285298351C37736160 @default.
- W4285298351 hasConceptScore W4285298351C41008148 @default.
- W4285298351 hasConceptScore W4285298351C45374587 @default.
- W4285298351 hasLocation W42852983511 @default.
- W4285298351 hasLocation W42852983512 @default.
- W4285298351 hasOpenAccess W4285298351 @default.
- W4285298351 hasPrimaryLocation W42852983511 @default.
- W4285298351 hasRelatedWork W2901368259 @default.
- W4285298351 hasRelatedWork W2903917280 @default.
- W4285298351 hasRelatedWork W2961085424 @default.
- W4285298351 hasRelatedWork W4205705013 @default.
- W4285298351 hasRelatedWork W4286629047 @default.
- W4285298351 hasRelatedWork W4306321456 @default.
- W4285298351 hasRelatedWork W4306674287 @default.
- W4285298351 hasRelatedWork W4319453497 @default.
- W4285298351 hasRelatedWork W4364383453 @default.
- W4285298351 hasRelatedWork W4224009465 @default.
- W4285298351 isParatext "false" @default.
- W4285298351 isRetracted "false" @default.
- W4285298351 workType "article" @default.