Plan First, Act Later: Streaming Reasoning for Executable Memory in Robot Control
Plan First, Act Later: Streaming Reasoning for Executable Memory in Robot Control
This research proposes PAL, a hierarchical memory framework for vision-language robot manipulation. It maintains a short-term visual buffer, semantic-event memory, and an executable subtask list. A high-level vision-language model converts streaming observations into subtasks, which are then executed by a low-level policy. PAL achieves 56.05% success on RoboMME and 60.83% in real-world tasks, outperforming existing baselines.
This paper introduces PAL, a hierarchical memory framework for vision-language robot manipulation, which improves long-horizon control by proactively writing subtasks into memory. The method outperforms existing baselines in both simulated and real-world tasks.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- code: not_reported
- code: Reported pending Open source S6
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
-dependent benchmarks and real-world manipulation tasks. On RoboMME, PAL achieves a \textbf{56.05\%} success rate, outperforming the strongest baseline, FrameSamp-Modul, at 44.51\%. In real-world experiments across four manipulation tasks, \pal{} reaches \textbf{60.83\%} overall success, surpassing MemER (45.00\%) and FrameSamp-Modul (28.33\%). These results show that proactively writing future-relevant subtasks into
Open source S5
hierarchical memory improves long-horizon visual reasoning and robot control. Code and videos can be found on our website: https://qinghangliu.github.io/pal-webpage/
Open source S6
Source：arXiv Robotics — research abstracts · arxiv.org