Search papers, labs, and topics across Lattice.
3
0
7
3
Achieve state-of-the-art robot manipulation success rates while slashing inference costs by up to 72% with A1, a fully open-source VLA framework that adaptively truncates computation.
By explicitly modeling part relationships with a graph, Graph-PiT generates images with significantly improved structural coherence compared to existing part-based image synthesis methods.
Current robot manipulation benchmarks fail to capture the messy reality of real-world deployment, so this work introduces a new benchmark, ManipArena, to close the sim2real gap.
