Dorsa Sadigh

EXPO: Stable Reinforcement Learning with Expressive Policies

Polychromic Objectives for Reinforcement Learning

Value Flows

What Matters for Batch Online Reinforcement Learning in Robotics?

SteerVLA: Steering Vision-Language-Action Models Toward Effective Long-Tail Driving

Steerable Video Action Model

Vision Language Models are In-Context Value Learners

Diffusion Models are Secretly Exchangeable: Parallelizing DDPMs via Auto Speculation

What's the Move? Hybrid Imitation Learning via Salient Points

Latent Diffusion Planning for Imitation Learning

Scaffolding Dexterous Manipulation with Vision-Language Models