FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3094476119> ?p ?o ?g. }

• W3094476119 abstract "Intent detection is one of the core components of goal-oriented dialog systems, and detecting out-of-scope (OOS) intents is also a practically important skill. Few-shot learning is attracting much attention to mitigate data scarcity, but OOS detection becomes even more challenging. In this paper, we present a simple yet effective approach, discriminative nearest neighbor classification with deep self-attention. Unlike softmax classifiers, we leverage BERT-style pairwise encoding to train a binary classifier that estimates the best matched training example for a user input. We propose to boost the discriminative ability by transferring a natural language inference (NLI) model. Our extensive experiments on a large-scale multi-domain intent detection task show that our method achieves more stable and accurate in-domain and OOS detection accuracy than RoBERTa-based classifiers and embedding-based nearest neighbor approaches. More notably, the NLI transfer enables our 10-shot model to perform competitively with 50-shot or even full-shot classifiers, while we can keep the inference time constant by leveraging a faster embedding retrieval model." @default.
• W3094476119 created "2020-10-29" @default.
• W3094476119 creator A5023164791 @default.
• W3094476119 creator A5032046813 @default.
• W3094476119 creator A5036357902 @default.
• W3094476119 creator A5054582135 @default.
• W3094476119 creator A5059955534 @default.
• W3094476119 creator A5066791810 @default.
• W3094476119 creator A5069021125 @default.
• W3094476119 creator A5090818171 @default.
• W3094476119 date "2020-10-24" @default.
• W3094476119 modified "2023-10-18" @default.
• W3094476119 title "Discriminative Nearest Neighbor Few-Shot Intent Detection by Transferring Natural Language Inference" @default.
• W3094476119 cites W1599016936 @default.
• W3094476119 cites W1815076433 @default.
• W3094476119 cites W1840435438 @default.
• W3094476119 cites W2115733720 @default.
• W3094476119 cites W2137291015 @default.
• W3094476119 cites W2141211166 @default.
• W3094476119 cites W2183341477 @default.
• W3094476119 cites W2251044566 @default.
• W3094476119 cites W2251869843 @default.
• W3094476119 cites W2531327146 @default.
• W3094476119 cites W2549975402 @default.
• W3094476119 cites W2556468274 @default.
• W3094476119 cites W2593864460 @default.
• W3094476119 cites W2601450892 @default.
• W3094476119 cites W2604763608 @default.
• W3094476119 cites W2604890851 @default.
• W3094476119 cites W2768282280 @default.
• W3094476119 cites W2786712888 @default.
• W3094476119 cites W2799010330 @default.
• W3094476119 cites W2891555348 @default.
• W3094476119 cites W2891844856 @default.
• W3094476119 cites W2922917409 @default.
• W3094476119 cites W2950541952 @default.
• W3094476119 cites W2950725899 @default.
• W3094476119 cites W2950729111 @default.
• W3094476119 cites W2951434086 @default.
• W3094476119 cites W2953109491 @default.
• W3094476119 cites W2953365054 @default.
• W3094476119 cites W2955315229 @default.
• W3094476119 cites W2962678612 @default.
• W3094476119 cites W2963310665 @default.
• W3094476119 cites W2963341924 @default.
• W3094476119 cites W2963341956 @default.
• W3094476119 cites W2963403868 @default.
• W3094476119 cites W2963564796 @default.
• W3094476119 cites W2963846996 @default.
• W3094476119 cites W2964212550 @default.
• W3094476119 cites W2965373594 @default.
• W3094476119 cites W2970139579 @default.
• W3094476119 cites W2970206392 @default.
• W3094476119 cites W2970871182 @default.
• W3094476119 cites W2971013448 @default.
• W3094476119 cites W2971048662 @default.
• W3094476119 cites W2971296908 @default.
• W3094476119 cites W2978017171 @default.
• W3094476119 cites W2988299267 @default.
• W3094476119 cites W2991468406 @default.
• W3094476119 cites W2995638926 @default.
• W3094476119 cites W2997771882 @default.
• W3094476119 cites W3012396850 @default.
• W3094476119 cites W3014333092 @default.
• W3094476119 cites W3016625483 @default.
• W3094476119 doi "https://doi.org/10.48550/arxiv.2010.13009" @default.
• W3094476119 hasPublicationYear "2020" @default.
• W3094476119 type Work @default.
• W3094476119 sameAs 3094476119 @default.
• W3094476119 citedByCount "16" @default.
• W3094476119 countsByYear W30944761192021 @default.
• W3094476119 countsByYear W30944761192022 @default.
• W3094476119 crossrefType "posted-content" @default.
• W3094476119 hasAuthorship W3094476119A5023164791 @default.
• W3094476119 hasAuthorship W3094476119A5032046813 @default.
• W3094476119 hasAuthorship W3094476119A5036357902 @default.
• W3094476119 hasAuthorship W3094476119A5054582135 @default.
• W3094476119 hasAuthorship W3094476119A5059955534 @default.
• W3094476119 hasAuthorship W3094476119A5066791810 @default.
• W3094476119 hasAuthorship W3094476119A5069021125 @default.
• W3094476119 hasAuthorship W3094476119A5090818171 @default.
• W3094476119 hasBestOaLocation W30944761191 @default.
• W3094476119 hasConcept C108583219 @default.
• W3094476119 hasConcept C113238511 @default.
• W3094476119 hasConcept C119857082 @default.
• W3094476119 hasConcept C153180895 @default.
• W3094476119 hasConcept C154945302 @default.
• W3094476119 hasConcept C188441871 @default.
• W3094476119 hasConcept C2776214188 @default.
• W3094476119 hasConcept C34736171 @default.
• W3094476119 hasConcept C41008148 @default.
• W3094476119 hasConcept C41608201 @default.
• W3094476119 hasConcept C774472 @default.
• W3094476119 hasConcept C95623464 @default.
• W3094476119 hasConcept C97931131 @default.
• W3094476119 hasConceptScore W3094476119C108583219 @default.
• W3094476119 hasConceptScore W3094476119C113238511 @default.
• W3094476119 hasConceptScore W3094476119C119857082 @default.
• W3094476119 hasConceptScore W3094476119C153180895 @default.
• W3094476119 hasConceptScore W3094476119C154945302 @default.

• next