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. 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. 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. 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. 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.
Figures & tables
Fig. 1: Illustration of Poly-Ripple . Each client m∈[M] performs ICL for the same server-issued query xq using its private demonstration set Pm . Under privacy and context length constraints, the server applies importance-aware aggregation to combine diverse client knowledge into the final prediction y^q .
Fig. 2: Raster scan order (left) is commonly used in visual autoregressive modeling [ 22 ] . In contrast, inspired by the physical law of ripple-like outward signal propagation from a Tx in indoor radio map estimation, we adopt a ripple order (right) to better capture the causal structure for next-token prediction.
Fig. 3: Each client m∈[M] locally performs ICL using an LAVM, which consists of a VQ-GAN encoder and decoder for tokenization and reconstruction, and a transformer for next-token prediction. The server aggregates the ripple-structured local predictions via A(⋅) to collaboratively produce the final radio map.
Fig. 4: Local prompt construction with K=3 on different clients.
Fig. 5: A query layout and client layouts with varying structural relevance.
Fig. 6: Aggregation weights under different temperature values τ .
Figure 7
Fig. 9: Qualitative comparison between uniform aggregation (avg) and the importance-aware ( αm ) aggregation used by Poly-Ripple .