Collected sources and patterns will appear here. Add from search, explore, or the patterns library.
Tuple[Corpus, Queries, DenseEncoder] -> RetrievalResults
Map query and document texts to vector embeddings and compute exact pairwise similarities to rank the documents.
Problem it solves
Retrieving documents requires scoring the entire corpus against a set of query representations in a shared vector space.
Consumes
Emits
The real projects this mechanism was found in. Attribution is the point — this is how the best teams actually do it.