Search papers, labs, and topics across Lattice.
2
0
4
8
Instead of struggling to generate multi-view consistent 3D edits, this paper cleverly uses reinforcement learning to *verify* consistency, unlocking high-quality 3D scene editing from 2D diffusion priors.
ImagineAgent's clever combination of cognitive maps and generative tools lets it crush previous state-of-the-art on OV-HOI tasks while needing only 20% of the training data.
