Research map / Graph learning / Knowledge and temporal graphs

Learning Continuously Evolving Spatio-Temporal Explanations for Traffic Flow Forecasting

NeurIPS 2026 · Graph learning › Knowledge and temporal graphs · approach: Spatio-temporal graph traffic forecasting

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes continuously evolving spatio-temporal explanations for traffic flow forecasting, moving beyond static or discrete feature attribution to interpretable explanations of model decisions.

Problem

Traffic flow forecasting models are black-box, and existing explainable methods rely on feature attribution that yields static or discrete evidence, failing to capture how spatio-temporal explanations evolve.

Approach

Shifts explanation from static/discrete feature attribution to continuously evolving spatio-temporal explanations that track how evidence changes over time and space within the forecasting process.

Closest papers by meaning

• Efficient Traffic Prediction Through Spatio-Temporal Distillation (AAAI 2025 · similarity 0.90)
• Spatiotemporal-aware Trend-Seasonality Decomposition Network for Traffic Flow Forecasting (AAAI 2025 · similarity 0.89)
• Towards Dynamic Spatial-Temporal Graph Learning: A Decoupled Perspective (AAAI 2024 · similarity 0.89)
• Decomposed Spatio-Temporal Mamba for Long-Term Traffic Prediction (AAAI 2025 · similarity 0.88)
• SSL-STMFormer Self-Supervised Learning Spatio-Temporal Entanglement Transformer for Traffic Flow Prediction (AAAI 2025 · similarity 0.88)
• Spatio-Temporal Pivotal Graph Neural Networks for Traffic Flow Forecasting (AAAI 2024 · similarity 0.87)