Introduces functional transparency to represent model predictions as sums of univariate and bivariate functions.
Explore technology by industry and role
Newest additions first. Explore past and current notes by industry and role.
All notes
Newest additions first. Explore past and current notes by industry and role.
Uses Crazyflie drone models to simulate spacecraft relative motion dynamics under linearized Clohessy-Wiltshire dynamics
Learned energy model CRAM solves reasoning tasks significantly faster than looping models
A new attribute—3D object visibility—is introduced to improve detection accuracy.
A sketching method reduces evaluation calls for context attribution by leveraging stability assumptions and backpropagation circuit differentiation
A simple yet effective method to enhance perception robustness against NeRF artifacts is proposed.
The Interaction Scene Graph effectively models interactions among ego-vehicle, road agents, and map elements by incorporating geometric priors.
A subject control mechanism enables diverse external data integration for scalable and varied data generation
Future task success is evaluated by comparing runs with and without memory segments, enabling cost-aware memory selection.
A multi-frame vision-language model that reduces memory usage by at least 10 times compared to existing models
AgentSSL proposes a method for MLE agents to combine and adapt established SSL techniques using unlabeled data.
A novel object discovery module based on spatial-temporal point cloud aggregation is proposed.
A novel game-based end-to-end learning framework is proposed to acquire human driving skills.
The first end-to-end roadside infrastructure system specifically designed to support autonomous driving, leveraging existing lampposts.
A modular platform based on the 2018 KIA Soul EV is introduced for autonomous driving research