Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
Affiliation:
2
0
4
2
Action chunking can be dynamically optimized using internal cross-attention signals, leading to significantly improved task success rates in VLA frameworks.
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.