Collaborative In-Context Learning for Indoor Radio Map Estimation

查看原文 / View Original 编辑 / Edit

Computer Science > Machine Learning arXiv:2610.31702 (cs) [Submitted on 21 Oct 2026] Title: Collaborative In-Context Learning for Indoor Radio Map Estimation Authors: Yuanzhe Peng, Jie Xu.

计算机科学 > 机器学习 arXiv:2610.31702 (cs) [提交于 2026 年 10 月 21 日] 标题：用于室内无线电地图估计的协作式上下文学习作者：Yuanzhe Peng, Jie Xu。

Abstract: Indoor radio map estimation (RME) is essential for wireless applications such as coverage planning, localization, and network deployment. However, most existing methods follow a supervised learning paradigm that requires costly labeled data and task-specific model training.

摘要：室内无线电地图估计（RME）对于覆盖规划、定位和网络部署等无线应用至关重要。然而，大多数现有方法遵循监督学习范式，需要昂贵的标注数据和特定任务的模型训练。

Moreover, indoor environments involve privacy-sensitive information, such as building layouts and structures, that often cannot be centralized. To address these challenges, we make the first attempt to formulate indoor RME as a collaborative in-context learning (ICL) problem.

此外，室内环境涉及建筑布局和结构等隐私敏感信息，这些信息通常无法集中处理。为了应对这些挑战，我们首次尝试将室内 RME 制定为一个协作式上下文学习（ICL）问题。

We propose Poly-Ripple, a framework in which multiple clients perform ICL using large autoregressive vision models (LAVMs) without model updates. Each client is prompted by its private demonstrations and predicts the radio map for the same scene query.

我们提出了 Poly-Ripple，这是一个框架，其中多个客户端使用大型自回归视觉模型（LAVM）执行 ICL，而无需更新模型。每个客户端通过其私有演示进行提示，并为相同的场景查询预测无线电地图。

To capture the causal structure of wireless propagation outward from the transmitter (Tx), we adopt ripple-ordered autoregressive modeling. We further introduce importance-aware aggregation to exploit heterogeneous contextual knowledge distributed across clients under the context length constraint.

为了捕捉从发射机（Tx）向外传播的无线传播因果结构，我们采用了涟漪排序自回归建模。我们进一步引入了重要性感知聚合，以在上下文长度约束下利用分布在各客户端的异构上下文知识。

We provide theoretical analysis and conduct extensive experiments on a public indoor RME benchmark. The experimental results show that Poly-Ripple consistently outperforms representative ICL baselines, while aligning with our theoretical analysis.

我们提供了理论分析，并在一个公开的室内 RME 基准测试上进行了广泛的实验。实验结果表明，Poly-Ripple 的表现始终优于代表性的 ICL 基线，并与我们的理论分析相一致。