Search papers, labs, and topics across Lattice.
5
0
8
6
Fine-tuned robotic policies can be biased towards certain instruction factors, but a new bias-aware data strategy can significantly enhance their performance with fewer demonstrations.
A single generalist model outperforms specialized systems, achieving over 35% improvement in real-world robotic task success.
Heterogeneous agents can achieve "mind reading" capabilities through dense alignment, outperforming traditional methods with significantly lower compute costs.
VoLoAgent outperforms traditional manipulation systems by seamlessly integrating planning, monitoring, and recovery in real-time, transforming how robots handle complex tasks in dynamic environments.
VLMs, despite strong spatial reasoning benchmark performance, exhibit a surprising and persistent bias: they confuse vertical position with distance, mirroring the perspective bias of natural photographs.
