Grounding
Also: grounded
Basing an answer on a named source or set of facts. A link to a source does not prove the answer follows from it.
The model answers from whatever it absorbed in training, like an exam from memory.
The application first finds the relevant pages in your documents and puts them in front of the model, like an open-book exam with page references.
Open book is better, but the student can still misread the page, and the search can hand it the wrong page. The page reference is what lets you check.
The fuller definition
Grounding connects an output to specified evidence, data, or an environment rather than relying solely on general learned patterns.
Go deeper
Delivery may be possible in the first week of April, subject to inspection.
Delivery was promised on 1 April.
The output changes certainty, invents a specific day, and loses a condition. A link to the real source makes checking possible; it does not make that summary accurate.
A concrete example
A summary is limited to supplied passages and identifies which passage supports each material statement.
The distinction that matters
Source-grounded generation can still omit contrary passages or overstate the available evidence.
Related terms
RAG — Retrieval-augmented generation
Letting the model look things up in your documents before answering, and show where the answer came from. Like an open-book exam.
Documents & retrievalCitation
A pointer to the source that supports a statement, so someone can check it.
Reliability & securityHallucination
When the model states something that sounds right but is made up, including cases, quotes, and facts that do not exist.
Where it is taught
How document-grounded AI answers workSource and further reading
Sentence Transformers · Semantic searchThe definition and example are written for this library. The link gives background, official documentation or legal authority as labelled. Legal scope depends on the jurisdiction and current text; product features depend on release and settings.