Σύγκριση μεθόδων
Εξετάστε τις επιλεγμένες μεθόδους δίπλα-δίπλα· οι γραμμές που διαφέρουν επισημαίνονται.
| Βελτιστοποίηση Άμεσης Προτίμησης× | QLoRA× | |
|---|---|---|
| Πεδίο | Βαθιά Μάθηση | Βαθιά Μάθηση |
| Οικογένεια | Machine learning | Machine learning |
| Έτος προέλευσης | 2023 | 2023 |
| Δημιουργός≠ | Rafael Rafailov | Tim Dettmers |
| Τύπος | Training methodology | Training methodology |
| Θεμελιώδης πηγή≠ | Rafailov, R., Sharma, A., Mitchell, E., Manning, C. D., Ermon, S., & Finn, C. (2023). Direct preference optimization: Your language model is secretly a reward model. arXiv preprint arXiv:2305.18290. link ↗ | Dettmers, T., Pagnoni, A., Holtzman, A., & Contrastive, L. (2023). QLoRA: Efficient finetuning of quantized LLMs. arXiv preprint arXiv:2305.14314. link ↗ |
| Εναλλακτικές ονομασίες | DPO, Direct preference | QLoRA, Quantized LoRA |
| Συναφείς | 4 | 4 |
| Σύνοψη≠ | Direct Preference Optimization (DPO) is a training method introduced by Rafailov et al. in 2023 that aligns language models with human preferences without requiring an explicit reward model. By directly optimizing for preference pairs (better response vs worse response), DPO simplifies the training pipeline compared to reinforcement learning from human feedback (RLHF). | 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. |
| ScholarGateΣύνολο δεδομένων ↗ |
|
|