expert trajectories
- Bigger, Regularized, Categorical: High-Capacity Value Functions are Efficient Multi-Task Learners
- DPAIL: Training Diffusion Policy for Adversarial Imitation Learning without Policy Optimization
- Learning from Demonstrations via Capability-Aware Goal Sampling
- Prioritizing Perception-Guided Self-Supervision: A New Paradigm for Causal Modeling in End-to-End Autonomous Driving