RLHF
Also known as: reinforcement learning from human feedback
Using human preference signals to align model outputs with policy and tone goals - common in modern LLM post-training stacks.
Contact if you need a term added for a security or procurement review.