FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3214447639> ?p ?o ?g. }

• W3214447639 abstract "With the wide availability of Pre-trained Language Models (PLMs), multi-task fine-tuning across domains has been extensively applied. For tasks related to distant domains with different class label sets, PLMs may memorize non-transferable knowledge for the target domain and suffer from negative transfer. Inspired by meta-learning, we propose the Meta Distant Transfer Learning (Meta-DTL) framework to learn the cross-task knowledge for PLM-based methods. Meta-DTL first employs task representation learning to mine implicit relations among multiple tasks and classes. Based on the results, it trains a PLM-based meta-learner to capture the transferable knowledge across tasks. The weighted maximum entropy regularizers are proposed to make meta-learner more task-agnostic and unbiased. Finally, the meta-learner can be fine-tuned to fit each task with better parameter initialization. We evaluate Meta-DTL using both BERT and ALBERT on seven public datasets. Experiment results confirm the superiority of Meta-DTL as it consistently outperforms strong baselines. We find that Meta-DTL is highly effective when very few data is available for the target task." @default.
• W3214447639 created "2021-11-22" @default.
• W3214447639 creator A5017227883 @default.
• W3214447639 creator A5028279831 @default.
• W3214447639 creator A5034474990 @default.
• W3214447639 creator A5054621636 @default.
• W3214447639 creator A5065197504 @default.
• W3214447639 creator A5066159151 @default.
• W3214447639 date "2021-01-01" @default.
• W3214447639 modified "2023-09-26" @default.
• W3214447639 title "Meta Distant Transfer Learning for Pre-trained Language Models" @default.
• W3214447639 cites W1986614398 @default.
• W3214447639 cites W2113459411 @default.
• W3214447639 cites W2135964261 @default.
• W3214447639 cites W2163302275 @default.
• W3214447639 cites W2165698076 @default.
• W3214447639 cites W2251939518 @default.
• W3214447639 cites W2601450892 @default.
• W3214447639 cites W2604474127 @default.
• W3214447639 cites W2604763608 @default.
• W3214447639 cites W2788496822 @default.
• W3214447639 cites W2804582864 @default.
• W3214447639 cites W2895531857 @default.
• W3214447639 cites W2951615109 @default.
• W3214447639 cites W2952841984 @default.
• W3214447639 cites W2962724755 @default.
• W3214447639 cites W2962739339 @default.
• W3214447639 cites W2962799101 @default.
• W3214447639 cites W2962897020 @default.
• W3214447639 cites W2963310665 @default.
• W3214447639 cites W2963341956 @default.
• W3214447639 cites W2963403868 @default.
• W3214447639 cites W2963846996 @default.
• W3214447639 cites W2963854351 @default.
• W3214447639 cites W2964078140 @default.
• W3214447639 cites W2964110616 @default.
• W3214447639 cites W2964914104 @default.
• W3214447639 cites W2970597249 @default.
• W3214447639 cites W2971167006 @default.
• W3214447639 cites W2980113592 @default.
• W3214447639 cites W2980708516 @default.
• W3214447639 cites W2984166938 @default.
• W3214447639 cites W2990704537 @default.
• W3214447639 cites W2995322030 @default.
• W3214447639 cites W2996428491 @default.
• W3214447639 cites W3011574394 @default.
• W3214447639 cites W3012255272 @default.
• W3214447639 cites W3034779619 @default.
• W3214447639 cites W3035127352 @default.
• W3214447639 cites W3035565536 @default.
• W3214447639 cites W3041133507 @default.
• W3214447639 cites W3041843309 @default.
• W3214447639 cites W3082274269 @default.
• W3214447639 cites W3098903812 @default.
• W3214447639 cites W3104613728 @default.
• W3214447639 cites W3104763958 @default.
• W3214447639 cites W3105470358 @default.
• W3214447639 cites W3170536409 @default.
• W3214447639 cites W3173256823 @default.
• W3214447639 cites W3100536164 @default.
• W3214447639 doi "https://doi.org/10.18653/v1/2021.emnlp-main.768" @default.
• W3214447639 hasPublicationYear "2021" @default.
• W3214447639 type Work @default.
• W3214447639 sameAs 3214447639 @default.
• W3214447639 citedByCount "1" @default.
• W3214447639 countsByYear W32144476392023 @default.
• W3214447639 crossrefType "proceedings-article" @default.
• W3214447639 hasAuthorship W3214447639A5017227883 @default.
• W3214447639 hasAuthorship W3214447639A5028279831 @default.
• W3214447639 hasAuthorship W3214447639A5034474990 @default.
• W3214447639 hasAuthorship W3214447639A5054621636 @default.
• W3214447639 hasAuthorship W3214447639A5065197504 @default.
• W3214447639 hasAuthorship W3214447639A5066159151 @default.
• W3214447639 hasBestOaLocation W32144476391 @default.
• W3214447639 hasConcept C114466953 @default.
• W3214447639 hasConcept C119857082 @default.
• W3214447639 hasConcept C137293760 @default.
• W3214447639 hasConcept C145420912 @default.
• W3214447639 hasConcept C150899416 @default.
• W3214447639 hasConcept C154945302 @default.
• W3214447639 hasConcept C162324750 @default.
• W3214447639 hasConcept C175154964 @default.
• W3214447639 hasConcept C187736073 @default.
• W3214447639 hasConcept C199360897 @default.
• W3214447639 hasConcept C204321447 @default.
• W3214447639 hasConcept C2780451532 @default.
• W3214447639 hasConcept C2781002164 @default.
• W3214447639 hasConcept C28006648 @default.
• W3214447639 hasConcept C30038468 @default.
• W3214447639 hasConcept C33923547 @default.
• W3214447639 hasConcept C41008148 @default.
• W3214447639 hasConceptScore W3214447639C114466953 @default.
• W3214447639 hasConceptScore W3214447639C119857082 @default.
• W3214447639 hasConceptScore W3214447639C137293760 @default.
• W3214447639 hasConceptScore W3214447639C145420912 @default.
• W3214447639 hasConceptScore W3214447639C150899416 @default.
• W3214447639 hasConceptScore W3214447639C154945302 @default.
• W3214447639 hasConceptScore W3214447639C162324750 @default.
• W3214447639 hasConceptScore W3214447639C175154964 @default.
• W3214447639 hasConceptScore W3214447639C187736073 @default.

• next