Technical Terms
RLHF (Reinforcement Learning from Human Feedback)
Definition
A training technique where human evaluators rank model outputs, and this feedback is used to train a reward model that guides the AI to produce more helpful, harmless, and honest responses. RLHF is a key technique behind ChatGPT and Claude.
In Plain English
Training the AI by having humans rate its answers as good or bad. The AI learns to give more answers like the ones humans liked and fewer answers like the ones they didn't. It's like training a dog with treats — rewarding good behaviour.