arXiv:2608.31167v2 Announce Type: replace-cross
Abstract: Model-based control can directly execute specified objectives, while learning can amortize such behaviors into reactive policies, making their combination a natural solution to multi-stage manipulation. We introduce Semantically UNified (SUN) Programs, typed executables that compile grounded relations into aligned optimal control objectives, satisfaction predicates, and learning rewards. Our harness, Kuafu, equips a foundation model as a task-level agent to orchestrate scene preparation, verification, residual RL, and data production. The agent uses program feedback to repair candidate programs and training diagnostics to calibrate relative reward weights, retaining accepted task semantics across tool calls. Across nine multi-stage manipulation tasks, Kuafu achieves 82.03% average success rate, significantly outperforming all learned baselines. Its learned controllers generate demonstrations at 10.57x the human-teleoperation rate, yielding data that improve visualpolicy success by 23.6 percentage points over the strongest baseline. The policies transfer zero-shot to physical Franka and Kinova robots, demonstrating sim-to-real generalization.
Abstract: Model-based control can directly execute specified objectives, while learning can amortize such behaviors into reactive policies, making their combination a natural solution to multi-stage manipulation. We introduce Semantically UNified (SUN) Programs, typed executables that compile grounded relations into aligned optimal control objectives, satisfaction predicates, and learning rewards. Our harness, Kuafu, equips a foundation model as a task-level agent to orchestrate scene preparation, verification, residual RL, and data production. The agent uses program feedback to repair candidate programs and training diagnostics to calibrate relative reward weights, retaining accepted task semantics across tool calls. Across nine multi-stage manipulation tasks, Kuafu achieves 82.03% average success rate, significantly outperforming all learned baselines. Its learned controllers generate demonstrations at 10.57x the human-teleoperation rate, yielding data that improve visualpolicy success by 23.6 percentage points over the strongest baseline. The policies transfer zero-shot to physical Franka and Kinova robots, demonstrating sim-to-real generalization.
This article has been indexed from cs.AI updates on arXiv.org
Read the original article: