FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3172642864> ?p ?o ?g. }

• W3172642864 abstract "When scaled to hundreds of billions of parameters, pretrained language models such as GPT-3 (Brown et al., 2020) achieve remarkable few-shot performance. However, enormous amounts of compute are required for training and applying such big models, resulting in a large carbon footprint and making it difficult for researchers and practitioners to use them. We show that performance similar to GPT-3 can be obtained with language models that are much “greener” in that their parameter count is several orders of magnitude smaller. This is achieved by converting textual inputs into cloze questions that contain a task description, combined with gradient-based optimization; exploiting unlabeled data gives further improvements. We identify key factors required for successful natural language understanding with small language models." @default.
• W3172642864 created "2021-06-22" @default.
• W3172642864 creator A5071144367 @default.
• W3172642864 creator A5076269589 @default.
• W3172642864 date "2021-01-01" @default.
• W3172642864 modified "2023-10-15" @default.
• W3172642864 title "It’s Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners" @default.
• W3172642864 cites W1599016936 @default.
• W3172642864 cites W1724438581 @default.
• W3172642864 cites W1821462560 @default.
• W3172642864 cites W2101210369 @default.
• W3172642864 cites W2111316763 @default.
• W3172642864 cites W2163568299 @default.
• W3172642864 cites W2169610 @default.
• W3172642864 cites W2403763698 @default.
• W3172642864 cites W2804897457 @default.
• W3172642864 cites W2805206884 @default.
• W3172642864 cites W2898662126 @default.
• W3172642864 cites W2899771611 @default.
• W3172642864 cites W2927746189 @default.
• W3172642864 cites W2946659172 @default.
• W3172642864 cites W2956105246 @default.
• W3172642864 cites W2963341956 @default.
• W3172642864 cites W2963403868 @default.
• W3172642864 cites W2963674932 @default.
• W3172642864 cites W2963809228 @default.
• W3172642864 cites W2965373594 @default.
• W3172642864 cites W2970161131 @default.
• W3172642864 cites W2970476646 @default.
• W3172642864 cites W2970597249 @default.
• W3172642864 cites W2978017171 @default.
• W3172642864 cites W2988975212 @default.
• W3172642864 cites W2989911337 @default.
• W3172642864 cites W2990704537 @default.
• W3172642864 cites W2994673210 @default.
• W3172642864 cites W2995335514 @default.
• W3172642864 cites W2996428491 @default.
• W3172642864 cites W2998554035 @default.
• W3172642864 cites W3004346089 @default.
• W3172642864 cites W3005700362 @default.
• W3172642864 cites W3015468748 @default.
• W3172642864 cites W3016263386 @default.
• W3172642864 cites W3026404337 @default.
• W3172642864 cites W3034292689 @default.
• W3172642864 cites W3034775979 @default.
• W3172642864 cites W3037312869 @default.
• W3172642864 cites W3082274269 @default.
• W3172642864 cites W3088059392 @default.
• W3172642864 cites W3098824823 @default.
• W3172642864 cites W3104738015 @default.
• W3172642864 cites W3105966348 @default.
• W3172642864 cites W3111372685 @default.
• W3172642864 cites W3119189390 @default.
• W3172642864 cites W3153427360 @default.
• W3172642864 cites W3177265267 @default.
• W3172642864 cites W3214897310 @default.
• W3172642864 cites W95183648 @default.
• W3172642864 cites W1905066734 @default.
• W3172642864 cites W2525127255 @default.
• W3172642864 doi "https://doi.org/10.18653/v1/2021.naacl-main.185" @default.
• W3172642864 hasPublicationYear "2021" @default.
• W3172642864 type Work @default.
• W3172642864 sameAs 3172642864 @default.
• W3172642864 citedByCount "187" @default.
• W3172642864 countsByYear W31726428642020 @default.
• W3172642864 countsByYear W31726428642021 @default.
• W3172642864 countsByYear W31726428642022 @default.
• W3172642864 countsByYear W31726428642023 @default.
• W3172642864 crossrefType "proceedings-article" @default.
• W3172642864 hasAuthorship W3172642864A5071144367 @default.
• W3172642864 hasAuthorship W3172642864A5076269589 @default.
• W3172642864 hasBestOaLocation W31726428641 @default.
• W3172642864 hasConcept C119857082 @default.
• W3172642864 hasConcept C127413603 @default.
• W3172642864 hasConcept C132943942 @default.
• W3172642864 hasConcept C137293760 @default.
• W3172642864 hasConcept C151730666 @default.
• W3172642864 hasConcept C154945302 @default.
• W3172642864 hasConcept C178790620 @default.
• W3172642864 hasConcept C185592680 @default.
• W3172642864 hasConcept C18903297 @default.
• W3172642864 hasConcept C195324797 @default.
• W3172642864 hasConcept C199360897 @default.
• W3172642864 hasConcept C201995342 @default.
• W3172642864 hasConcept C204321447 @default.
• W3172642864 hasConcept C26517878 @default.
• W3172642864 hasConcept C2778344882 @default.
• W3172642864 hasConcept C2779439875 @default.
• W3172642864 hasConcept C2780451532 @default.
• W3172642864 hasConcept C2780936489 @default.
• W3172642864 hasConcept C38652104 @default.
• W3172642864 hasConcept C41008148 @default.
• W3172642864 hasConcept C47737302 @default.
• W3172642864 hasConcept C74912251 @default.
• W3172642864 hasConcept C86803240 @default.
• W3172642864 hasConceptScore W3172642864C119857082 @default.
• W3172642864 hasConceptScore W3172642864C127413603 @default.
• W3172642864 hasConceptScore W3172642864C132943942 @default.
• W3172642864 hasConceptScore W3172642864C137293760 @default.
• W3172642864 hasConceptScore W3172642864C151730666 @default.

• next