AHA-WAM:Asynchronous Horizon-Adaptive World-Action Modeling with Observation-Guided Context Routing
- Published
- Source
- arXiv
- Paper number
- 374
- Field
- Robotics
- arXiv ID
- 2606.09811
Key points
- The design separates Video DiT as a low-frequency planner and Action DiT as a high-frequency executor through an asynchronous decomposition.
- Rolling KV memory preserves the planner's temporal context over time.
- OVCR dynamically routes and updates the planner context based on the latest observations.
- Horizon-adaptive offset training learns arbitrary phase relationships between planner and executor.
- The method reaches SOTA-level performance, with 92.80 percent on RoboTwin and 78.3 percent success in the real world.
- It achieves control frequencies up to 56.9 Hz, which is a 10.82x speedup over Fast-WAM.
Paper links
External research summaries. These are not HDATF publications or measured product results.