Search papers, labs, and topics across Lattice.
DeepCybo
6
0
6
Human-as-Humanoid achieves a staggering 4.8–7.2x increase in demonstration throughput, transforming how humanoid robots learn from human actions.
G2Rec captures user interest prototypes more accurately than existing methods, enabling generative recommendation systems to operate without ground-truth user interests.
Achieving 3.8x higher recall with a co-designed framework that integrates graph construction, representation learning, and real-time serving could redefine large-scale recommendation systems.
KVEraser achieves a 3-4x speedup over full recomputation while maintaining high performance in long-context tasks, revolutionizing how we handle context updates in LLMs.
Tactile learning can now leverage a novel dataset and alignment method to dramatically enhance vision-tactile representation, outperforming traditional approaches.
AetheRock not only streamlines data collection for robot learning but also pairs it with a novel framework that enhances tactile learning in the face of sensor inconsistencies.
