H-GAR: A Hierarchical Interaction Framework via Goal-Driven Observation-Action Refinement for Robotic Manipulation

作者：Yijie Zhu, Rui Shao, Ziyang Liu, Jie He, Jizhihui Liu, Jiuru Wang, Zitong Yu · 单位：Harbin Institute of Technology, Shenzhen, Great Bay University, Shenzhen Loop Area Institute, Linyi University, Dongguan Key Laboratory for Intelligence and Information Technology · 会议/期刊：AAAI 2026 · 日期：2025-11-21 · 来源：Low-Level Learning-Based Action Modelling / Input Modelling / 2D Non-LLM-based Vision Language Action Models