RAPID: Robot Agentic Programming from Demonstrations
RAPID turns one visual demonstration into an executable, testable robot program, showing how agent loops can derive their own specifications, tools, and verification environment.
RAPID infers a task specification, robot action primitives, and an execution environment from **one visual demonstration**, then generates, verifies, and refines a program.
The reusable pattern for coding-agent builders is the closed loop: derive executable tests and tools from an example, run the program, and refine it against observed behavior. Its object-centric representation also separates strategy from scene-specific geometry.
RAPID infers a task specification, robot action primitives, and an execution environment from **one visual demonstration**, then generates, verifies, and refines a program. The reusable pattern for coding-agent builders is the closed loop: derive executable tests and tools from an example, run the program, and refine it against observed behavior. Its object-centric representation also separates strategy from scene-specific geometry. The evidence covers simulation and a Franka arm across **eight contact-rich tasks**, plus LIBERO-Pro evaluations. The abstract reports generalization across several variations but provides no comparative rates.
RAPID extends example-driven agent programming into physical manipulation: one demonstration supplies enough structure to infer tools, tests, and an executable program, while observed execution closes the refinement loop. It reinforces verification-centered harnesses and sharpens the abstraction boundary between semantic strategy and embodiment-specific control, but its limited robotics evidence does not establish general coding-agent reliability.