SADP: Subgoal-Aware Diffusion Policy for Long-Horizon Manipulation Learned from Foundation Model Generated Demonstrations

arXiv:2605.16871 · cs.RO · Submitted 2026-05-16 · Read on arXiv

cs.RO

Submitted: 2026-05-16

Updated: 2026-10-17

Comments: Author-accepted manuscript; published in IEEE Robotics and Automation Letters

Journal ref: IEEE Robotics and Automation Letters. VOL. 11, NO. 11, NOVEMBER 2026

DOI: 10.1109/LRA.2026.3742285

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Foundation Model — 2× in this paper · explained in 3 episodes
• Subgoal-Aware Diffusion Policy — 2× in this paper
• Foundation Models — 1× in this paper · explained in 7 episodes
• Interpretability — 1× in this paper · explained in 15 episodes

Related papers

• FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
• MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
• RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
• ARC: A Reasoning Recipe for Robot Foundation Models
• HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
• APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies