SpatCode: Rotary-based Unified Encoding Framework for Efficient Spatiotemporal Vector Retrieval

TL;DR

SpatCode introduces a unified spatiotemporal vector retrieval framework that cohesively encodes semantic, temporal, and geographic cues into a single embedding space. It leverages a Rotary-based Unified Encoding to map time and location into unit-norm vectors, a Circular Incremental Update mechanism for efficient streaming updates, and a Weighted Interest-based Retrieval algorithm to enable context-aware, per-query modality weighting. Extensive experiments on diverse datasets show that SpatCode achieves higher recall with lower latency than state-of-the-art baselines and maintains robustness under dynamic data evolution. The approach provides a practical, scalable solution for real-time, multimodal spatiotemporal retrieval in intelligent systems.

Abstract

Spatiotemporal vector retrieval has emerged as a critical paradigm in modern information retrieval, enabling efficient access to massive, heterogeneous data that evolve over both time and space. However, existing spatiotemporal retrieval methods are often extensions of conventional vector search systems that rely on external filters or specialized indices to incorporate temporal and spatial constraints, leading to inefficiency, architectural complexity, and limited flexibility in handling heterogeneous modalities. To overcome these challenges, we present a unified spatiotemporal vector retrieval framework that integrates temporal, spatial, and semantic cues within a coherent similarity space while maintaining scalability and adaptability to continuous data streams. Specifically, we propose (1) a Rotary-based Unified Encoding Method that embeds time and location into rotational position vectors for consistent spatiotemporal representation; (2) a Circular Incremental Update Mechanism that supports efficient sliding-window updates without global re-encoding or index reconstruction; and (3) a Weighted Interest-based Retrieval Algorithm that adaptively balances modality weights for context-aware and personalized retrieval. Extensive experiments across multiple real-world datasets demonstrate that our framework substantially outperforms state-of-the-art baselines in both retrieval accuracy and efficiency, while maintaining robustness under dynamic data evolution. These results highlight the effectiveness and practicality of the proposed approach for scalable spatiotemporal information retrieval in intelligent systems.

Figures (6)

• Figure 1: Advantages of SpatCode Framework. Traditional spatiotemporal vector retrieval methods rely on additional filters for time and location, increasing system complexity and reducing efficiency. In contrast, unified encoding (SpatCode) embeds spatiotemporal cues into a unified representation, enabling efficient retrieval via a single inner product.
• Figure 2: SpatCode framework. SpatCode comprises a Rotary-based Unified Encoding Method for integrating spatiotemporal and heterogeneous vectors, a Circular Incremental Update Mechanism for efficient streaming updates, and a Weighted Interest-based Retrieval Algorithm for adaptive, personalized search.
• Figure 3: Per-record insertion and query latency over thirteen months in the sliding-window.
• Figure 4: Recall@10 of SpatCode with or without weighted interest-based retrieval.
• Figure 5: Query latency versus recall@10 of SpatCode at various HNSW $e_f$.