Search papers, labs, and topics across Lattice.
1
0
3
Standard retriever evaluations hide critical weaknesses in agentic search systems, but a new benchmark and training method exposes and addresses these flaws.
