Search papers, labs, and topics across Lattice.
4
0
8
2
The KITScenes Multimodal dataset redefines the landscape of autonomous driving research with the most complete HD maps and high-fidelity sensor data available publicly.
Explicitly modeling human-object interactions boosts multi-person human mesh recovery accuracy by up to 9.9%, showing that interaction context is key to understanding human pose and shape in complex scenes.
Current multimodal models can't handle the rapid-fire tactical analysis required for boxing commentary, as revealed by a new dataset and evaluation framework.
Training data is not enough: reasoning traces from diverse cultural backgrounds are critical for safe and reliable autonomous driving in rare, long-tail scenarios.
