Uporedite metode
Pregledajte izabrane metode jednu pored druge; redovi koji se razlikuju su istaknuti.
| GPT fino-podešavanje× | LoRA i PEFT× | Vision Transformer× | |
|---|---|---|---|
| Oblast | Duboko učenje | Duboko učenje | Duboko učenje |
| Porodica | Machine learning | Machine learning | Machine learning |
| Godina nastanka≠ | 2019 | 2022 | 2021 |
| Tvorac≠ | Radford, A. et al. (OpenAI) | Hu, E. J. et al.; Lester, B. et al. | Dosovitskiy, A. et al. |
| Tip≠ | Fine-tuning of pretrained autoregressive language models | Parameter-efficient fine-tuning of large pretrained models | Transformer architecture for images (self-attention over patches) |
| Temeljni izvor≠ | Radford, A., Wu, J., Child, R., Luan, D., Amodei, D. & Sutskever, I. (2019). Language Models are Unsupervised Multitask Learners. OpenAI Technical Report. link ↗ | Hu, E. J. et al. (2022). LoRA: Low-Rank Adaptation of Large Language Models. ICLR. link ↗ | Dosovitskiy, A. et al. (2021). An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. ICLR. link ↗ |
| Drugi nazivi≠ | GPT İnce Ayar ve Talimat Uyarlaması, GPT fine-tuning, instruction tuning, LLM fine-tuning | LoRA ve PEFT — Parametre Verimli İnce Ayar, Low-Rank Adaptation, parameter-efficient fine-tuning, prefix tuning | Görsel Transformer (ViT), görsel transformer, ViT, patch transformer for images |
| Srodne | 5 | 5 | 5 |
| Sažetak≠ | GPT fine-tuning adapts pretrained autoregressive language models such as GPT-2/3/4 or LLaMA — introduced in OpenAI's 2019 work by Radford and colleagues — to domain-specific data or to instruction following via reinforcement learning from human feedback (RLHF) or DPO. It is used for instruction following, domain adaptation, and generative tasks. | LoRA (Low-Rank Adaptation), introduced by Hu et al. in 2022, and the broader family of parameter-efficient fine-tuning (PEFT) methods adapt large pretrained language models to new tasks by training only a small number of extra parameters instead of every weight in the model. This makes fine-tuning possible with far less GPU memory and compute while leaving the original model largely untouched. | The Vision Transformer (ViT), introduced by Dosovitskiy and colleagues in 2021, splits an image into fixed-size patches, treats those patches as a sequence, and applies the Transformer self-attention mechanism to image classification. Given enough training data, it surpasses convolutional neural networks (CNNs). |
| ScholarGateSkup podataka ↗ |
|
|
|