Research
For a decision-making system to operate in the real world, its decisions must be interpretable to people and physically valid. We work along these axes.
Reliable Long-Horizon Behavior Synthesis
Keeping generated plans physically valid, beyond the horizon of the training data.
- Manifold-aware diffusion planning
- Trajectory stitching for long-horizon plans
- Detecting and refining infeasible plans
Physical AI
Decision making for embodied agents that act in the physical world.
- Manipulation and locomotion
- Unsupervised skill discovery via curriculum RL
Explainable AI
Interpreting the criteria behind a policy’s actions before execution.
- Feature attribution along faithful paths
- Interpretable deployment of learned policies
- Counterfactual activation editing
World Models
Learning predictive models of environment dynamics that agents can plan with.
- Action-conditioned future prediction with video diffusion
- Planning and policy learning inside learned world models
Multi-Agent AI
Coordinating multiple agents so that individually learned behaviors compose into safe teamwork.
- Multi-robot motion planning with diffusion models
- Scaling single-agent skills to team-level coordination