Search papers, labs, and topics across Lattice.
5
0
7
0
Confidence-guided active learning can drastically reduce localized errors in embodied world models, leading to more reliable predictions in complex environments.
Achieving a 25% reduction in prediction error, MOSH-WM redefines the landscape of object-centric video forecasting by grounding state representations in visual support.
Verified execution experience can be transformed into a reusable resource, boosting model performance on complex workflows by up to 15.5 percentage points.
AutoResearch achieves a 1.85-point improvement in mean Recall while reducing audit-confirmed issues, showcasing a new standard for reliability in autonomous research systems.
Forget random exploration: this framework uses affordance graphs and self-evolution to create coherent, complex robot tasks in simulation, leading to better policy learning and generalization.
