arXiv cs.AI· Changchuan Yang, Haoxuan Xu, Wenbo Chen, Shuai Ren, Jianlong Zheng, Huarui Zhang, Tianfu Li, Guanzhong Tian·· 9 天前AI 评分22
PMTRM：面向具身策略学习的伪记忆时间重编码模块
PMTRM: Pseudo-Memory Temporal Re-encoding Module for Embodied Policy Learning
AI 导读
研究者提出伪记忆时间重编码模块 PMTRM，仅 7.61M 参数，可将已执行状态与动作的有界历史编码为隐序列，供现有策略使用，以缓解机器人操作中局部观测相似但阶段动作不同的相位歧义。该模块通过时间异质性目标惩罚序列远距离位置的正相似度，并用锚点与重建损失保留动作预测所需信息，重建解码器仅用于训练。在仿真与真实机器人上搭配多种策略骨干实验，相位歧义任务成功率提升且额外计算开销很小。
正文 · 原文
Abstract:Robotic manipulation often contains repeated motions whose local observations look similar at different phases. When these phases require different actions, a policy that relies mainly on the current observation may repeat completed motions or switch phases at the wrong time. To address this phase ambiguity, we present the Pseudo-Memory Temporal Re-encoding Module (PMTRM), a lightweight plug-in module with only 7.61M parameters that encodes a bounded history of executed states and actions into a latent sequence for existing policies. To help distinguish phases, a temporal heterogeneity objective penalizes positive similarity between distant positions in this sequence, while anchor and reconstruction losses preserve information needed for action prediction. The reconstruction decoder is used only during training, leaving the temporal re-encoder to supply history to the policy at inference. We train the module progressively on synthetic sequences and robot data, then jointly with the policy, using temporal masking to accommodate partial histories. This integration retains the original action head and action space and adds auxiliary losses to the original policy loss. Experiments with multiple policy backbones in simulation and on a real robot show improved task success on tasks with phase ambiguity, with little additional computation.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.11168 [cs.RO] |
| (or arXiv:2610.11168v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.11168 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Changchuan Yang [view email]
[v1]
Thu, 8 Oct 2026 03:25:12 UTC (1,539 KB)
来源：arXiv cs.AI · arxiv.org