Data/ML/AIHigh signalEmergingAround since 2021
LoRA / PEFT
Also written as LoRA, Low-Rank Adaptation, PEFT, Parameter-Efficient Fine-Tuning
Techniques for adapting a large model by training a small number of extra parameters instead of retraining the whole thing — far cheaper, and the usual way fine-tuning is actually done.
Think of it like
Tailoring an off-the-peg suit rather than weaving new cloth.
Junior or senior?
Distinguishes people who have genuinely fine-tuned from people who have read about it.
Senior sounds like
Can say why they fine-tuned rather than prompting or using retrieval.
Ask them
“Why did fine-tuning beat prompting or retrieval for that problem?”