Search papers, labs, and topics across Lattice.
Shanghai Jiao Tong University
8
0
5
DeformGen transforms the landscape of deformable manipulation by enabling effective policy learning through innovative state augmentation and trajectory adaptation techniques.
LightSTAR slashes retrieval latency while maintaining top-tier accuracy by cleverly bypassing heavy MLLM processing on every document page.
Shifting the focus from marginal probabilities to joint trajectory probabilities, dVLA-RL achieves unprecedented success rates in robotic manipulation tasks.
ImageWAM shows that image editing can outperform video generation in robot action prediction, cutting costs and improving efficiency.
Learning dexterous manipulation from monocular videos could redefine how robots acquire complex skills without costly teleoperation data.
AHA-WAM achieves a remarkable 92.80% success rate on RoboTwin while executing actions at 24.17 Hz, all without the need for prior robot-data training.
Achieving 86.4% grasp stability and 83.3% real-world success, SynManDex bridges the gap between human dexterity and robotic manipulation.
Imagine a self-improving sensor network that anticipates where the most valuable data will be, leading to dramatically better field reconstructions even with sparse measurements.
