Search papers, labs, and topics across Lattice.
6
0
9
VAIC enables humanoid robots to perform complex tasks in real-world settings without the need for perfect state observability, significantly advancing their practical deployment.
Explicitly leveraging structural context in entity alignment can significantly enhance performance, with ContextEA outperforming fine-tuned models on unseen knowledge graphs.
Forget global coordinates: this new method unlocks long-context and streaming 3D reconstruction by predicting relative constraints between frames.
By explicitly aligning visuomotor representations with the action space's rigid-body geometry, OASIS achieves superior robotic manipulation performance compared to methods relying on implicit geometry recovery.
Achieve competitive novel-view synthesis with a 4MB footprint and 78ms inference by learning a compact global scene representation *before* decoding 3D geometry.
Generating robotic manipulation data that respects object affordances is now possible at scale, but current imitation learning methods still struggle with tasks like pouring and hanging, revealing a critical gap.
