Search papers, labs, and topics across Lattice.
Affiliation:
8
0
10
15
Collision-free motion planning just got a major upgrade with NeurRAFT, which reshapes trajectory distributions to prioritize safety without sacrificing performance.
Imagined tactile observations can boost robotic manipulation performance by over 44% without any physical tactile sensors.
MoRE achieves a staggering 44 percentage point increase in deployment success rates by seamlessly integrating behavior mode redirection into policy weights, eliminating the need for inference-time adjustments.
Achieving up to 46% token compression without sacrificing accuracy, HMPO revolutionizes the efficiency of chain-of-thought reasoning in large language models.
End-to-end robot motion planning gets a boost: Flow Motion Policy leverages flow matching to generate diverse, collision-free paths, outperforming traditional methods in success rate and efficiency.
Contact-rich robotic manipulation gets a 14% performance boost by adaptively ignoring force/torque signals when they're irrelevant, and fusing them intelligently with vision when contact matters.
Tactile force fields unlock robust robotic disassembly, outperforming vision-only and even naive RGB-tactile fusion, particularly in deformable and contact-rich scenarios.
Foundation model-powered robots need more than just physical constraints—think modular safety guardrails that understand context, human intent, and evolving norms.
