Search papers, labs, and topics across Lattice.
3
0
8
Unlock scalable CAD generation from unannotated 3D meshes with DreamCAD, a framework that directly produces editable BREPs from point-level supervision, outperforming existing methods and achieving over 75% user preference.
By stitching together the best reasoning steps from multiple diffusion-generated trajectories, this method achieves significant accuracy gains and latency reductions in math and coding tasks, outperforming traditional diffusion models and unified architectures.
Current LLM agents are surprisingly bad at synthesizing information from multiple sources to solve realistic problems, achieving dismal scores on the new DEEPSYNTH benchmark.
