LLMs
LoRA (Low-Rank Adaptation)
Freezing pretrained W and learning ΔW = (α/r)·BA with B initialised to zero, training ~0.1–2% of parameters.
Why interviewers ask about it
B = 0 means training starts exactly at the pretrained model. Targeting all linear layers matters more than increasing rank - most tutorials get this wrong.
Related terms
This term is part of the free AI/ML Engineer interview preparation module - browse the full glossary for every definition.