Collected sources and patterns will appear here. Add from search or the patterns library.
(Query, VectorIndex, SparseIndex) -> RankedList<Chunk>
Combine keyword-based and vector-based search candidates, then re-rank them using a cross-encoder model.
Problem it solves
Single-mode search misses either lexical exact matches or semantic concepts.
Consumes
Emits
The real projects this mechanism was found in. Attribution is the point — this is how the best teams actually do it.