Krahasoni metodat

	QLoRA	Auto-koduesit e maskuar
Fusha	Mësimi i thellë	Mësimi i thellë
Familja	Machine learning	Machine learning
Viti i origjinës	2023	2021
Krijuesi	Tim Dettmers	Kaiming He
Lloji	Training methodology	Neural network architecture
Burimi themelues	Dettmers, T., Pagnoni, A., Holtzman, A., & Contrastive, L. (2023). QLoRA: Efficient finetuning of quantized LLMs. arXiv preprint arXiv:2305.14314.	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).
Emërtime të tjera	QLoRA, Quantized LoRA	MAE, Vision MAE
Të lidhura	4	4
Përmbledhja	QLoRA 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.