Glossary
Groundedness
On this page
The short answer
- Groundedness is the habit of tying answers to supplied sources rather than to the model’s prior knowledge. The word is often used alongside faithfulness; some tools name a “groundedness” feedback function that scores this directly.
- This is a glossary entry, not a product claim or a benchmark result.
- All terms · Templates · Guides.
Definition
Groundedness is the habit of tying answers to supplied sources rather than to the model’s prior knowledge. The word is often used alongside faithfulness; some tools name a “groundedness” feedback function that scores this directly.
In practice
If your app is supposed to answer only from a knowledge base, treat ungrounded but true answers as failures — they will go wrong the first time the knowledge base disagrees with the model.
The Riverbank Library RAG template includes a bad retrieved flyer (P4) that must not be used, which is a groundedness test as well as a faithfulness test.
Related terms
FAQ
Does Litmiq implement “Groundedness” as a built-in metric?
Only where the definition itself says so. Many terms are industry vocabulary. Litmiq is a single-turn, BYOK prompt tester with an LLM judge, not a full evaluation platform.
Where is the rest of the glossary?
The hub at /glossary lists every term. This page is the standalone entry for Groundedness.
See the term in a suite
Many glossary entries have a matching template you can run in the browser.
- Free, no signup
- Runs in your browser
- Your OpenAI key, sent only to OpenAI