RFW-A: Agentic Procedural Generation of 400K Articulated Assets over 500 Categories

High-quality articulated assets are essential for embodied perception, simulation, and manipulation, yet constructing them at scale remains difficult. Manual anno- tation and procedural modeling require substantial human effort. Recent agentic systems automate asset construction through program synthesis, but repeat costly reasoning and refinement for every new asset. Our key observation is that as- sets within a category can share construction logic despite variations in structure and appearance. Agentic synthesis and procedural generation are therefore com- plementary: agents can construct reusable programs, while procedural execution reuses the resulting logic across many samples. We therefore present RFW-A, which combines the two in a new articulated-asset generation paradigm. Start- ing from existing articulated programs, RFW-A organizes reusable components and their dependencies into category-level generators. Agents test and refine these generators across sampled configurations and motion states. Once validated, the generators produce diverse assets through procedural execution without further model inference, at an illustrative amortized API cost of $0.035 per asset under the stated accounting assumptions, about 1/32.6 of the per-attempt API cost re- ported for Articraft. Using RFW-A, we built RFW-A-400K, containing over 400K sim-ready articulated assets across more than 500 categories. Experiments show high structural and kinematic validity and stable execution across physics simulators. Beyond asset validity, training with RFW-A-400K improves part- level articulation estimation on seen and unseen categories, while policies trained in simulation with retrieved and program-edited assets transfer zero-shot to real articulated objects.