Search papers, labs, and topics across Lattice.
4
0
4
7
Attention-level generalization in Pelican-VLA 0.5 allows it to focus on relevant objects without any task-specific training, outperforming traditional models in unseen scenarios.
IOI achieves state-of-the-art simulation performance by decoupling deterministic motion from stochastic physical interactions, enabling robust zero-shot generalization to unseen tasks.
Current world models fail to evolve independently of observation, often freezing events instead of allowing them to progress unseen.
VLA models may excel at visually grounded tasks, but VLA-Trace reveals they still struggle with fine-grained semantic understanding and exhibit distinct modality processing strategies.
