Search papers, labs, and topics across Lattice.
4
0
8
4
Current action-conditioned world models fail to reliably follow diverse off-expert actions, risking the effectiveness of policy learning in real-world applications.
Achieving over 10% improvement in manipulation success rates, Lift3D-VLA redefines the integration of 3D geometry and action generation in robotic systems.
Achieving up to 4.7x speedup in TGNN training without sacrificing accuracy could redefine performance benchmarks in dynamic graph applications.
Visual grounding in VLAs weakens in deeper layers, but injecting multi-level visual features and pruning irrelevant tokens can boost performance by 9% in simulation and 7.5% in the real world.
