Search papers, labs, and topics across Lattice.
2
0
3
0
FlashVector is an agentic system that optimizes performance across all layers of the model serving stack, and the key contribution is an extensible framework to generalize the single kernel optimization agent paradigm to heterogeneous technical stacks, and to deliver performance improvements holistically.
Re-ranking control alone boosts key performance metrics by over 2%, but extending it to fine ranking unlocks even greater gains without sacrificing system stability.
