Grounding
Grounding is constraining a model to answer only from supplied source material, rather than from its training data.
A grounded prompt says explicitly that the answer must come from the provided documents and that the model should say so when the documents do not contain the answer. Without that second clause, a grounded prompt still invents an answer when the source is silent.
Related terms
- RAG (retrieval-augmented generation)
- RAG is a technique that retrieves relevant documents from an external store and places them in the prompt, so the model answers from that material rather than from memory.
- Hallucination
- A hallucination is model output that is presented confidently as fact but is fabricated — an invented citation, statistic, quotation, function name, or event.
- Citation
- A citation is a reference in model output pointing to the specific source passage that supports a claim.
More on failure modes
See the full glossary, read the guides, or put it into practice in the prompt builder.