FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3045492832> ?p ?o ?g. }

• W3045492832 abstract "Building conversational systems in new domains and with added functionality requires resource-efficient models that work under low-data regimes (i.e., in few-shot setups). Motivated by these requirements, we introduce intent detection methods backed by pretrained dual sentence encoders such as USE and ConveRT. We demonstrate the usefulness and wide applicability of the proposed intent detectors, showing that: 1) they outperform intent detectors based on fine-tuning the full BERT-Large model or using BERT as a fixed black-box encoder on three diverse intent detection data sets; 2) the gains are especially pronounced in few-shot setups (i.e., with only 10 or 30 annotated examples per intent); 3) our intent detectors can be trained in a matter of minutes on a single CPU; and 4) they are stable across different hyperparameter settings. In hope of facilitating and democratizing research focused on intention detection, we release our code, as well as a new challenging single-domain intent detection dataset comprising 13,083 annotated examples over 77 intents." @default.
• W3045492832 created "2020-08-03" @default.
• W3045492832 creator A5014866912 @default.
• W3045492832 creator A5045402928 @default.
• W3045492832 creator A5049308833 @default.
• W3045492832 creator A5064527875 @default.
• W3045492832 creator A5078083355 @default.
• W3045492832 date "2020-01-01" @default.
• W3045492832 modified "2023-10-14" @default.
• W3045492832 title "Efficient Intent Detection with Dual Sentence Encoders" @default.
• W3045492832 cites W2077302143 @default.
• W3045492832 cites W2084142050 @default.
• W3045492832 cites W2095705004 @default.
• W3045492832 cites W2413533759 @default.
• W3045492832 cites W2604698497 @default.
• W3045492832 cites W2611049140 @default.
• W3045492832 cites W2786860129 @default.
• W3045492832 cites W2791169651 @default.
• W3045492832 cites W2794557536 @default.
• W3045492832 cites W2889798024 @default.
• W3045492832 cites W2891732163 @default.
• W3045492832 cites W2948110372 @default.
• W3045492832 cites W2951216772 @default.
• W3045492832 cites W2953958347 @default.
• W3045492832 cites W2961915345 @default.
• W3045492832 cites W2962739339 @default.
• W3045492832 cites W2963341956 @default.
• W3045492832 cites W2963809228 @default.
• W3045492832 cites W2965373594 @default.
• W3045492832 cites W2966087730 @default.
• W3045492832 cites W2969574947 @default.
• W3045492832 cites W2970352191 @default.
• W3045492832 cites W2970618241 @default.
• W3045492832 cites W2970887634 @default.
• W3045492832 cites W2971160427 @default.
• W3045492832 cites W2972437240 @default.
• W3045492832 cites W2978017171 @default.
• W3045492832 cites W2997771882 @default.
• W3045492832 cites W3104078590 @default.
• W3045492832 cites W3177265267 @default.
• W3045492832 doi "https://doi.org/10.18653/v1/2020.nlp4convai-1.5" @default.
• W3045492832 hasPublicationYear "2020" @default.
• W3045492832 type Work @default.
• W3045492832 sameAs 3045492832 @default.
• W3045492832 citedByCount "92" @default.
• W3045492832 countsByYear W30454928322019 @default.
• W3045492832 countsByYear W30454928322020 @default.
• W3045492832 countsByYear W30454928322021 @default.
• W3045492832 countsByYear W30454928322022 @default.
• W3045492832 countsByYear W30454928322023 @default.
• W3045492832 crossrefType "proceedings-article" @default.
• W3045492832 hasAuthorship W3045492832A5014866912 @default.
• W3045492832 hasAuthorship W3045492832A5045402928 @default.
• W3045492832 hasAuthorship W3045492832A5049308833 @default.
• W3045492832 hasAuthorship W3045492832A5064527875 @default.
• W3045492832 hasAuthorship W3045492832A5078083355 @default.
• W3045492832 hasBestOaLocation W30454928321 @default.
• W3045492832 hasConcept C111919701 @default.
• W3045492832 hasConcept C118505674 @default.
• W3045492832 hasConcept C119857082 @default.
• W3045492832 hasConcept C124952713 @default.
• W3045492832 hasConcept C134306372 @default.
• W3045492832 hasConcept C142362112 @default.
• W3045492832 hasConcept C154945302 @default.
• W3045492832 hasConcept C178790620 @default.
• W3045492832 hasConcept C185592680 @default.
• W3045492832 hasConcept C204321447 @default.
• W3045492832 hasConcept C2777530160 @default.
• W3045492832 hasConcept C2778344882 @default.
• W3045492832 hasConcept C2780980858 @default.
• W3045492832 hasConcept C33923547 @default.
• W3045492832 hasConcept C36503486 @default.
• W3045492832 hasConcept C41008148 @default.
• W3045492832 hasConcept C76155785 @default.
• W3045492832 hasConcept C8642999 @default.
• W3045492832 hasConcept C94915269 @default.
• W3045492832 hasConceptScore W3045492832C111919701 @default.
• W3045492832 hasConceptScore W3045492832C118505674 @default.
• W3045492832 hasConceptScore W3045492832C119857082 @default.
• W3045492832 hasConceptScore W3045492832C124952713 @default.
• W3045492832 hasConceptScore W3045492832C134306372 @default.
• W3045492832 hasConceptScore W3045492832C142362112 @default.
• W3045492832 hasConceptScore W3045492832C154945302 @default.
• W3045492832 hasConceptScore W3045492832C178790620 @default.
• W3045492832 hasConceptScore W3045492832C185592680 @default.
• W3045492832 hasConceptScore W3045492832C204321447 @default.
• W3045492832 hasConceptScore W3045492832C2777530160 @default.
• W3045492832 hasConceptScore W3045492832C2778344882 @default.
• W3045492832 hasConceptScore W3045492832C2780980858 @default.
• W3045492832 hasConceptScore W3045492832C33923547 @default.
• W3045492832 hasConceptScore W3045492832C36503486 @default.
• W3045492832 hasConceptScore W3045492832C41008148 @default.
• W3045492832 hasConceptScore W3045492832C76155785 @default.
• W3045492832 hasConceptScore W3045492832C8642999 @default.
• W3045492832 hasConceptScore W3045492832C94915269 @default.
• W3045492832 hasLocation W30454928321 @default.
• W3045492832 hasLocation W30454928322 @default.
• W3045492832 hasLocation W30454928323 @default.
• W3045492832 hasOpenAccess W3045492832 @default.
• W3045492832 hasPrimaryLocation W30454928321 @default.

• next