论文/arXiv 2610.11085·20 天前

DaCe-DT：面向异构任务的基于自适应提示与轨迹校正的数据中心离线多任务强化学习

针对异构离线多任务强化学习的数据质量与分布问题，DaCe-DT通过自适应提示和轨迹校正，在Meta-World上验证了稳健学习效果。

AI 整理

• 本体 SDK 与控制
• 数据与采集
• 具身模型

原文链接：DaCe-DT: Data-Centric Offline Multi-Task Reinforcement Learning via Adaptive Prompts and Trajectory Correction for Heterogeneous Tasks（arXiv 2610.11085 · 发布日期：2026-10-08）

DaCe-DT: Data-Centric Offline Multi-Task Reinforcement Learning via Adaptive Prompts and Trajectory Correction for Heterogeneous Tasks

针对异构离线多任务强化学习的数据质量与分布问题，DaCe-DT通过自适应提示和轨迹校正，在Meta-World上验证了稳健学习效果。

核心要点

• 离线MTRL受预收集数据质量与分布制约，现有方法较少改进数据层面。
• 揭示提示长度利用不足、随机提示语义无关和碎片轨迹误导监督三类瓶颈。
• 提出LGPM、RAPC与VARC，实现数据中心的提示适配和轨迹细化。
• 在Meta-World上较SOTA在最优/次优数据集平均提升11.73%/13.34%。

验证范围与局限：验证范围为Meta-World上的最优与次优离线数据集。

英文摘要 (Abstract)

Offline multi-task reinforcement learning (Offline MTRL) heavily depends on the quality and distribution of pre-collected data. However, existing methods mainly focus on algorithmic optimization, with less emphasis on data-level improvements to enhance learning ability and generalization performance. This paper, from a data perspective, reveals three key bottlenecks that limit Offline MTRL performance:(i) ineffective utilization of prompts length under diverse task complexities, and (ii) semantic irrelevance of randomly sampled prompt segments, (iii) misleading supervision induced by fragmented and discontinuous trajectories. To address these challenges, we propose DaCe-DT, a robust offline MTRL framework designed to be insensitive to heterogeneous task complexities and data quality, featuring length-gated prompt masking (LGPM), retrieval-augmented prompt construction (RAPC), and value-adaptive return calibration (VARC). Together, these mechanisms enable DaCe-DT to deliver data-centric prompt adaptation and trajectory refinement, resulting in robust multi-task generalization and stable policy learning amid heterogeneous offline data and tasks. Experimental results on Meta-World show that DaCe-DT consistently outperforms state-of-the-art methods, achieving an average improvement of 11.73% on optimal datasets and an improvement of 13.34% on suboptimal datasets, demonstrating its effectiveness in learning stably from imperfect data and improving overall multi-task performance.

作者：Xinfei Wang, Shanchen Pang, Chenhao Zhang, Shudong Wang, Wenhao Ji, Haiyuan Gui, Meng Han, Xiaojian Liao

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.11085：DaCe-DT: Data-Centric Offline Multi-Task Reinforcement Learning via Adaptive Prompts and Trajectory Correction for Heterogeneous Tasks访问 2026-10-21