Just for FUNS: LLM-Guided Spatio-Temporal Graph Node Generation for Forecasting Unobserved Node States

Computer Science > Machine Learning arXiv:2610.08818 (cs) [Submitted on 24 Sep 2026 (v1), last revised 8 Oct 2026 (this version, v2)] Title: Just for FUNS: LLM-Guided Spatio-Temporal Graph Node Generation for Forecasting Unobserved Node States Authors: Shuhao Li, Weidong Yang, Changan Liu, Wei Zhuo, Yingbo Zhou, Fan Zhang, Siqiang Luo.

计算机科学 > 机器学习 arXiv:2610.08818 (cs) [提交于 2026 年 9 月 24 日 (v1)，最后修订于 2026 年 10 月 8 日 (此版本，v2)] 标题：Just for FUNS：用于预测未观测节点状态的 LLM 引导时空图节点生成。作者：Shuhao Li, Weidong Yang, Changan Liu, Wei Zhuo, Yingbo Zhou, Fan Zhang, Siqiang Luo。

Abstract: Spatio-temporal forecasting is a cornerstone of logistics, urban planning, and intelligent transportation systems. However, constrained by deployment costs and maintenance resources, sensor networks often lack comprehensive spatial coverage, rendering Forecast Unobserved Node States (FUNS) a critical yet formidable challenge. Conventional models rely on historical observations and typically falter when encountering nodes without prior records.

摘要：时空预测是物流、城市规划和智能交通系统的基石。然而，受限于部署成本和维护资源，传感器网络往往缺乏全面的空间覆盖，使得“预测未观测节点状态”（FUNS）成为一项关键而艰巨的挑战。传统模型依赖于历史观测数据，在遇到没有先前记录的节点时通常会失效。

To address this, we redefine the problem as a conditional generation task on spatio-temporal graphs and propose GenST, a framework that introduces Large Language Models (LLMs) as a semantic bridge, leveraging a pre-trained LLM fine-tuned to extract rich semantic features from node descriptions, such as functional zones and road network structures, to compensate for missing spatio-temporal signals.

为了解决这一问题，我们将该问题重新定义为时空图上的条件生成任务，并提出了 GenST 框架。该框架引入大语言模型（LLMs）作为语义桥梁，利用经过微调的预训练 LLM 从节点描述（如功能区和路网结构）中提取丰富的语义特征，以补偿缺失的时空信号。

Specifically, we design a two-stage generative architecture: a Spatio-Temporal VAE first compresses spatio-temporal dynamics into a latent space, followed by a Generative Transformer (GenT) that reconstructs the future states of unobserved nodes from noise, guided by multi-modal conditions including semantics, geographic coordinates, and neighborhood contexts.

具体而言，我们设计了一种两阶段生成架构：时空变分自编码器（Spatio-Temporal VAE）首先将时空动态压缩到潜在空间中，随后由生成式 Transformer（GenT）在语义、地理坐标和邻域上下文等多模态条件的引导下，从噪声中重构未观测节点的未来状态。

Experiments on six traffic and two non-traffic datasets show GenST significantly outperforms existing baselines in zero-shot prediction tasks, demonstrating the practical potential of semantic-guided generation for mitigating spatio-temporal data sparsity.

在六个交通数据集和两个非交通数据集上的实验表明，GenST 在零样本预测任务中显著优于现有的基准模型，证明了语义引导生成在缓解时空数据稀疏性方面的实际潜力。