Search papers, labs, and topics across Lattice.
Stanford University
1
0
2
Get the performance boost of expensive sampling-based RL policies for a fraction of the compute by learning to prune action candidates early in the diffusion denoising process.
