Research map / Vision-language and multimodal / Robot action and navigation
H-GAR: A Hierarchical Interaction Framework via Goal-Driven Observation-Action Refinement for Robotic Manipulation
Yijie Zhu, Rui Shao, Ziyang Liu, Jie He, Jizhihui Liu, Jiuru Wang, Zitong Yu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add GOS (goal-conditioned observation synthesis) (ablation) | H-GAR baseline · PushT | success rate: 0.90 -> 0.92 | Improved |
| add GOS (ablation) | H-GAR baseline · PushT-M | success rate: 0.78 -> 0.82 | Improved |
| add GOS (ablation) | H-GAR baseline · Libero-10 | success rate: 0.85 -> 0.89 | Improved |
| add IAAR without memory bank (ablation) | H-GAR with GOS · Libero-10 | success rate: 0.89 -> 0.91 | Improved |
| add historical action memory bank to IAAR (ablation) | H-GAR with GOS+IAAR · PushT | success rate: 0.96 -> 0.99 | Improved |
| add historical action memory bank to IAAR (ablation) | H-GAR with GOS+IAAR · PushT-M | success rate: 0.87 -> 0.90 | Improved |
| add historical action memory bank to IAAR (ablation) | H-GAR with GOS+IAAR · Libero-10 | success rate: 0.91 -> 0.94 | Improved |
| use goal frame only as GOS input instead of uniform multi-frame (ablation) | H-GAR · PushT-M | success rate: 0.83 -> 0.90 | Improved |
| use goal frame only instead of single random frame (ablation) | H-GAR · Libero-10 | success rate: 0.88 -> 0.94 | Improved |
| use similarity-based memory update instead of random (ablation) | H-GAR · Libero-10 | success rate: 0.89 -> 0.94 | Improved |
| use similarity-based memory update instead of FIFO (ablation) | H-GAR · PushT | success rate: 0.97 -> 0.99 | Improved |
| increase memory bank size from 32 to 64 (ablation) | H-GAR · Libero-10 | success rate: 0.94 -> 0.92 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CoT-VLA — CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models (CVPR 2025) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- PIVOT-R: Primitive-Driven Waypoint-Aware World Model for Robotic Manipulation (NeurIPS 2024 · similarity 0.90)
- SAM2Act: Integrating Visual Foundation Model with A Memory Architecture for Robotic Manipulation (ICML 2025 · similarity 0.89)
- Chain-of-Action: Trajectory Autoregressive Modeling for Robotic Manipulation (NeurIPS 2025 · similarity 0.89)
- Predictive Inverse Dynamics Models are Scalable Learners for Robotic Manipulation (ICLR 2025 · similarity 0.88)
- Beyond World-Frame Action Heads: Motion-Centric Action Frames for Vision-Language-Action Models (NeurIPS 2026 · similarity 0.88)
- $\textit{Hyper-GoalNet}$: Goal-Conditioned Manipulation Policy Learning with HyperNetworks (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
