Grounding is the technique of anchoring a language model's responses in verifiable, real-world data rather than relying solely on its training knowledge. RAG is the most common grounding method: the model retrieves relevant documents before answering. Grounding reduces hallucinations, keeps answers current, and provides citations. A Moroccan legal assistant grounded in the official journal can quote exact articles, while an ungrounded one might invent plausible-sounding but wrong legal references.
NLP
Grounding
Related terms
Learn to use these concepts in practice.
Join the 212AY Academy →