Search papers, labs, and topics across Lattice.
5
0
7
3
LMMs can slash FLOPs by 89% without sacrificing accuracy, thanks to a frequency-modulated visual restoration technique that preserves crucial visual semantics even with fewer tokens.
A single meta-RL policy can now handle 66% mass variations and 70% rotor thrust losses in quadrotors, achieving zero-shot sim-to-real transfer for agile maneuvers.
Achieve the seemingly impossible: ASTER uses RL to enable cable-suspended quadrotors to perform autonomous inverted flight.
Even the best LLMs struggle to maintain correct intermediate states when solving university-level STEM problems, often taking more steps than necessary and accumulating errors along the way.
Achieve time-optimal MAV flight at 18 m/s in cluttered environments by combining imitation learning with model predictive contouring control.
