Search papers, labs, and topics across Lattice.
1
0
2
5
Shifting reasoning to the indexing stage can drastically reduce query latency while enhancing retrieval effectiveness through LLM-generated rationales.