Search papers, labs, and topics across Lattice.
3
0
4
Overcoming the pitfalls of proposal overgeneration, ASAR achieves up to a 28% boost in event completion success by leveraging adjacent action sequences.
Latent world models can be salvaged with a simple trick: learn a reachability metric on logged trajectories to fix the planner's terminal cost function.
Injecting "beneficial noise" into cross-attention mechanisms can significantly improve unsupervised domain adaptation by forcing models to focus on content rather than style distractions.
