Rekod bukti kaedah

QLoRA

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.

Sources recorded, not reviewed

Rekod sumber

Petikan disalin secara verbatim daripada rekod sumber kaedah. Tiada pengesahan peringkat tuntutan disimpulkan daripadanya.

Efficient Finetuning of Quantized LLMs

Rekod kaedah taksonomik · ml-model / deep-learning

Dettmers, T., Pagnoni, A., Holtzman, A., & Contrastive, L. (2023). QLoRA: Efficient finetuning of quantized LLMs. arXiv preprint arXiv:2305.14314. · URL

Buka kaedah penuh

Tuntutan yang dikurasi

Tuntutan disimpan dalam lejar bukti, setiap satu dengan penilaiannya sendiri.

Tiada tuntutan terkurasi lagi

Pandangan ini tidak mencipta penilaian tuntutan apabila lejar tiada.

Kaedah berkaitan

Dijana daripada graf kaedah dan ditunjukkan sebagai perhubungan yang dicadangkan mesin — tiada tuntutan bukti disimpulkan.

Same method familyDirect Preference Optimizationmachine-suggested · Relational suggestion, not evidence.Same method familyLatent Diffusion Modelsmachine-suggested · Relational suggestion, not evidence.Same method familyMamba (State Space Model)machine-suggested · Relational suggestion, not evidence.Same method familyMasked Autoencodersmachine-suggested · Relational suggestion, not evidence.