Search papers, labs, and topics across Lattice.
Rutgers University
4
0
6
6
MobiDiff achieves a 5.3x speedup in generating synthetic mobility data while maintaining high fidelity to real-world patterns.
Achieve 9.97% higher accuracy in cross-domain human activity recognition while simultaneously reducing computation by 6.4x with a new sensor data tokenization and attention mechanism.
Forget tedious human-in-the-loop DAgger – WM-DAgger uses World Models to synthesize corrective actions for robots, achieving impressive 93% success with just 5 demos.
Forget two-stage depth completion pipelines: Any2Full achieves state-of-the-art results with a single-stage prompting approach that's both faster and more accurate.
