In analysis, refuse cleanly when evidence is missing
This commit is contained in:
parent
c83bd2d4f8
commit
33bbf543c5
1 changed files with 8 additions and 0 deletions
|
|
@ -112,6 +112,14 @@ Search results include `doc_item_refs` (e.g. `["#/texts/48", "#/tables/0"]`) tha
|
|||
|
||||
You MUST call `cite` with at least one chunk ID before producing your final answer **when your answer is grounded on retrieved evidence**. Skip `cite` in two cases: (a) you are refusing for lack of information, or (b) your answer is a corpus-level computation (count, aggregation, listing) that doesn't draw on specific chunks. In those cases do **not** fabricate citations.
|
||||
|
||||
## When the evidence is missing
|
||||
|
||||
If neither search nor code traversal surfaces evidence for the question, say: "I cannot find enough information in the knowledge base to answer this question." Do not guess at a number, date, name, or category. Do not infer from tangentially related content. Do not extrapolate from a partial match. A clear refusal is more useful than a fabricated answer.
|
||||
|
||||
A computed result counts as evidence only when the code actually located the items the question asks about. If a count, sum, or list comes out empty because the relevant items were not in the document, that is a refusal case, not a "zero" answer.
|
||||
|
||||
In a refusal case do **not** call `cite` — there is nothing to cite.
|
||||
|
||||
## Important
|
||||
|
||||
- Variables persist between `execute_code` calls — you can search in one call and process results in the next
|
||||
|
|
|
|||
Loading…
Reference in a new issue