arXiv cs.AI· Joseph Natter, Long Vu, Yifan Zhang, Bobin Deng, Kun Suo, Rui Wu·· 13 小时前AI 评分31
GAS-PFN：面向零样本交通预测的图感知时空先验数据拟合网络
GAS-PFN: Graph-Aware Spatiotemporal Prior-Data Fitted Networks for Zero-Shot Traffic Forecasting
AI 导读
研究者提出图感知时空先验数据拟合网络 GAS-PFN，通过随机游走扩散将采样道路拓扑注入合成任务先验，使模型在预训练中学习图耦合动态，并借助共享 Transformer 主干的双掩码推理分别建模传感器时序与邻域时空交互，无需参数更新即可适配目标路网。在 90 分钟预测跨 6 个基准上，其 MSE 较前代 TimePFN 平均降低 34.0%，较四个近期业界通用零样本预测器平均降低 27.7%。
正文 · 原文
Abstract:Traffic forecasting requires modeling complex temporal dynamics and the spatial dependencies imposed by road topology. Supervised graph-based models achieve strong performance, but are often trained for a particular dataset and sensor graph, making deployment to new or evolving networks costly. Prior-Data Fitted Networks (PFNs) offer a zero-shot alternative by learning in-context inference from synthetic task priors, which provide an explicit mechanism for encoding domain-specific assumptions. Existing forecasting PFNs, however, do not explicitly represent the graph-structured dependencies governing traffic dynamics. To combine graph-aware traffic modeling with zero-shot adaptation, we introduce the Graph-Aware Spatiotemporal Prior-Data Fitted Network (GAS-PFN). GAS-PFN incorporates sampled road topology into its synthetic task prior through random-walk diffusion, teaching the model to learn graph-coupled dynamics during pretraining. Its co-designed inference architecture uses two mask-conditioned passes through a shared Transformer backbone to separately model sensor-specific temporal dynamics and spatiotemporal interactions among neighboring sensors before integrating their representations for prediction. This design enables GAS-PFN to use target-network structure and context without parameter updates. Across six benchmarks at the 90-minute horizon, GAS-PFN reduces MSE by 34.0% on average relative to TimePFN, its closest PFN-based predecessor. It also reduces MSE by 27.7% on average relative to four recent industry-developed general-purpose zero-shot forecasters. In this comparison, all models receive 90 minutes of contiguous recent history, and any additional context is drawn from each dataset's chronological training split. These results indicate that aligning PFN priors and inference architectures with graph structure improves in-context learning for zero-shot traffic forecasting.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE) |
| Cite as: | arXiv:2610.37598 [cs.LG] |
| (or arXiv:2610.37598v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.37598 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Joseph Natter [view email]
[v1]
Tue, 27 Oct 2026 17:31:26 UTC (1,667 KB)
来源：arXiv cs.AI · arxiv.org