NVIDIA paper AW-0 jointly generates skill code, future video frames and robot actions, reaching 50.7% average success on RoboTwin 2.0
"Agentic World Action Models"
What if robots could write their own code to plan actions, while also predicting how the physical world will respond?
NVIDIA's new AW-0 combines three capabilities in one model: generating reusable skill programs, predicting future video frames, and producing robot actions. All three are generated together, which enables code-based planning to guide physical execution.
On RoboTwin 2.0, AW-0 achieves 50.7% average success, compared to 41.3% for π0.5. Across eight real-world manipulation tasks, it scores 71.1% vs 46.0% for the strongest baseline, including tasks like folding towels and solving a Rubik's Cube.
alphaxiv.org/abs/2610.30136XSources2
- @askalphaxivX"Agentic World Action Models" What if robots could write their own code to plan actions, while also predicting how the physical world will respond? NVIDIA's new AW-0 combines three capabilities in one model: generating reusable skill programs, predicting future video frames, and producing robot actions. All three are generated together, which enables code-based planning to guide physical execution. On RoboTwin 2.0, AW-0 achieves 50.7% average success, compared to 41.3% for π0.5. Across eight real-world manipulation tasks, it scores 71.1% vs 46.0% for the strongest baseline, including tasks like folding towels and solving a Rubik's Cube. https://www.alphaxiv.org/abs/2610.30136View original ·
arXivResearchAgentic World Action ModelsView original ·
Headline by dalao.ai. The quoted post is the record; its author keeps authorship.