haiku.rag/tests/multi_db
Yiorgis Gozadinos d9f489dcc8
Deduplicate search results within one model response
Sibling searches emitted in one response overlap heavily (40.6% of
returned chunk slots on Glimmer ORB fan-out cases). A result whose
rendered evidence a sibling already showed keeps its rank slot but
collapses to a reference line, and a picture attaches once per response
keyed on (source, document_id, self_ref). Equivalence is the
format_for_agent rendering at neutral rank/total plus picture keys,
bucketed under the qualified chunk id, so another database's copy or a
different expansion of the same anchor formats in full.

Search state now commits only after formatting and image construction
succeed: a raising image build no longer leaves results citable that
the model never saw, notes evidence for them, or suppresses a later
sibling.
2026-09-02 12:29:12 +03:00
..
__init__.py Split the multi-database tests by subject 2026-08-26 13:43:53 +03:00
conftest.py Finish the comment pass, and escape document fields everywhere Rich renders 2026-08-28 15:34:47 +03:00
helpers.py State what the code does, not what it replaced 2026-08-28 15:13:52 +03:00
test_capabilities.py Deduplicate search results within one model response 2026-09-02 12:29:12 +03:00
test_citations.py Price qa.max_searches in search units 2026-09-02 12:29:12 +03:00
test_documents.py State what the code does, not what it replaced 2026-08-28 15:13:52 +03:00
test_expansion.py Make three tests assert what they are about 2026-08-27 18:04:28 +03:00
test_lifecycle.py Finish the comment pass, and escape document fields everywhere Rich renders 2026-08-28 15:34:47 +03:00
test_scope.py Finish the comment pass, and escape document fields everywhere Rich renders 2026-08-28 15:34:47 +03:00
test_search.py Order cross-database fusion by cosine similarity to the query 2026-09-01 15:26:42 +03:00