ScholarGate
Assistant

Comparer des méthodes

Examinez les méthodes sélectionnées côte à côte ; les lignes qui diffèrent sont mises en évidence.

QLoRA×Autoencodeurs masqués×
DomaineApprentissage profondApprentissage profond
FamilleMachine learningMachine learning
Année d'origine20232021
Auteur d'origineTim DettmersKaiming He
TypeTraining methodologyNeural network architecture
Source fondatriceDettmers, T., Pagnoni, A., Holtzman, A., & Contrastive, L. (2023). QLoRA: Efficient finetuning of quantized LLMs. arXiv preprint arXiv:2305.14314. link ↗He, K., Chen, X., Xie, S., Li, Y., Dollár, P., & Girshick, R. (2022). Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 16000-16009). DOI ↗
AliasQLoRA, Quantized LoRAMAE, Vision MAE
Apparentées44
RésuméQLoRA is an efficient fine-tuning method introduced by Dettmers et al. in 2023 that enables fine-tuning large language models using quantization and low-rank adaptation. By combining 4-bit quantization with LoRA, QLoRA reduces memory requirements by 75%, enabling fine-tuning of 65B-parameter models on single GPUs.Masked Autoencoders (MAE) is a self-supervised learning approach introduced by He et al. in 2021 that masks random patches of an image and trains a model to reconstruct the missing content. Adapting the masked language modeling paradigm from NLP to vision, MAE learns rich visual representations by solving a challenging reconstruction task without requiring labels.
ScholarGateJeu de données
  1. v1
  2. 1 Sources
  3. PUBLISHED
  1. v1
  2. 1 Sources
  3. PUBLISHED

Aller à la recherche Télécharger les diapositives

ScholarGateComparer des méthodes: QLoRA · Masked Autoencoders. Consulté le 2026-06-18 sur https://scholargate.app/fr/compare