PEFT (e.g., LoRA, T-Few) updates a small subset of parameters (often new ones) using labeled, task-specific data, unlike classic fine-tuning, which updates all parameters—Option A is correct. Option B reverses PEFT’s efficiency. Option C (no modification) fits soft prompting, not all PEFT. Option D (all parameters) mimics classic fine-tuning. PEFT reduces resource demands.
OCI 2025 Generative AI documentation likely contrasts PEFT and fine-tuning under customization methods.
Contribute your Thoughts:
Chosen Answer:
This is a voting comment (?). You can switch to a simple comment. It is better to Upvote an existing comment if you don't have anything to add.
Submit