ScholarGate
Asystent

Porównaj metody

Przeglądaj wybrane metody obok siebie; wiersze, które się różnią, są wyróżnione.

Destylacja wiedzy×Longformer / BigBird×
DziedzinaUczenie głębokieUczenie głębokie
RodzinaMachine learningMachine learning
Rok powstania20152020
TwórcaHinton, G., Vinyals, O. & Dean, J.Beltagy, Peters & Cohan (Longformer); Zaheer et al. (BigBird)
TypNeural network compression (teacher–student)Sparse-attention Transformer for long sequences
Źródło pierwotneHinton, G., Vinyals, O. & Dean, J. (2015). Distilling the Knowledge in a Neural Network. NeurIPS Deep Learning Workshop. link ↗Beltagy, I., Peters, M. E. & Cohan, A. (2020). Longformer: The Long-Document Transformer. arXiv. link ↗
Inne nazwyBilgi Damıtma (Knowledge Distillation), bilgi damıtma, teacher-student distillation, model distillationUzun Dizi Transformer (Longformer / BigBird), uzun dizi transformer, long-document transformer, sparse-attention transformer
Pokrewne54
PodsumowanieKnowledge Distillation is a model-compression technique, introduced by Geoffrey Hinton and colleagues in 2015, that trains a small student model using the soft-label outputs of a large teacher model. Distilled models such as DistilBERT and TinyBERT reach roughly 97% of the larger model's performance while running far faster.Long-sequence Transformers such as Longformer (Beltagy, Peters & Cohan, 2020) and BigBird (Zaheer et al., 2020) replace the standard Transformer's O(n²) attention with sparse attention patterns that scale linearly, O(n), with sequence length. This lets a single model attend over thousands of tokens — full documents, legal texts, or genomic sequences — that would not fit a conventional Transformer.
ScholarGateZbiór danych
  1. v1
  2. 2 Źródła
  3. PUBLISHED
  1. v1
  2. 2 Źródła
  3. PUBLISHED

Przejdź do wyszukiwania Pobierz slajdy

ScholarGatePorównaj metody: Knowledge Distillation · Longformer / BigBird. Pobrano 2026-06-18 z https://scholargate.app/pl/compare