Search papers, labs, and topics across Lattice.
Affiliation:
3
0
5
Ride-hailing dispatchers do not need brittle reward engineering: training a preference reward model directly on implicit multi-view passenger and driver trajectories outperforms heuristic-tuned hold policies in live, city-scale deployment.
RAW restoration consistently beats generic RGB models, but ISP-aware RGB models take the crown for best performance.
