Search papers, labs, and topics across Lattice.
6
0
7
5
DenseReward synthesizes diverse failure trajectories automatically, enabling robots to learn from a rich array of failure modes without human labeling.
VisualClaw slashes API costs by 98% while boosting accuracy, transforming how VLMs can operate in real-time environments.
Robots can now adjust their manipulation speed on-the-fly, achieving both rapid execution in low-risk phases and precision in high-risk tasks.
The hardest AI tasks remain largely unsolved, with current models achieving only a 2.6% success rate on economically valuable workflows.
Ditch the clunky tool-use pipelines: STORM teaches video-language models to reason about space and time using *internalized* latent trajectories, slashing inference costs while boosting accuracy.
GUI agents can achieve significantly stronger task-solving capabilities through carefully designed post-training and data curation, without relying on costly online data collection.
