Research map / Graph learning / Knowledge and temporal graphs
Revisiting Spatial-Temporal Similarity: A Deep Learning Framework for Traffic Prediction
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add flow gating and periodically shifted attention | DMVST-Net · NYC-Taxi | Start RMSE: 25.74±0.26 -> 24.10±0.25*** | Helped |
| Add flow gating and periodically shifted attention | DMVST-Net · NYC-Bike | Start RMSE: 9.14±0.13 -> 8.85±0.11*** | Helped |
| Add flow gating and periodically shifted attention | DMVST-Net · NYC-Taxi | Start MAPE: 17.38±0.46% -> 16.30±0.23%*** | Helped |
| Add flow gating and periodically shifted attention | DMVST-Net · NYC-Bike | Start MAPE: 22.20±0.33% -> 21.84±0.36%** | Helped |
| Add flow gating and periodically shifted attention | DMVST-Net · NYC-Taxi | End RMSE: 20.51±0.46 -> 19.05±0.31*** | Helped |
| Add flow gating and periodically shifted attention | DMVST-Net · NYC-Bike | End RMSE: 8.50±0.19 -> 8.15±0.15*** | Helped |
| Add flow gating and periodically shifted attention | DMVST-Net · NYC-Taxi | End MAPE: 17.14±0.32% -> 16.25±0.26%*** | Helped |
| Add flow gating and periodically shifted attention | DMVST-Net · NYC-Bike | End MAPE: 21.56±0.49% -> 20.87±0.39%*** | Helped |
| Add flow gating instead of using flow only as features (ablation) | LSTN-FI · NYC-Taxi and NYC-Bike | RMSE and MAPE | Helped |
| Add flow gating to local spatial-temporal modeling (ablation) | LSTN · NYC-Taxi and NYC-Bike | RMSE and MAPE | Helped |
| Add long-term information by concatenating it with short-term information in one LSTM (ablation) | LSTN · NYC-Taxi and NYC-Bike | RMSE and MAPE | No clear effect |
| Process long-term and short-term dependencies with separate LSTMs (ablation) | LSTN-L · NYC-Taxi and NYC-Bike | RMSE and MAPE | Helped |
Closest papers by meaning
- Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting (AAAI 2019 · similarity 0.90)
- TESTAM: A Time-Enhanced Spatio-Temporal Attention Model with Mixture of Experts (ICLR 2024 · similarity 0.90)
- Decomposed Spatio-Temporal Mamba for Long-Term Traffic Prediction (AAAI 2025 · similarity 0.89)
- Towards Dynamic Spatial-Temporal Graph Learning: A Decoupled Perspective (AAAI 2024 · similarity 0.88)
- Spatiotemporal-aware Trend-Seasonality Decomposition Network for Traffic Flow Forecasting (AAAI 2025 · similarity 0.88)
- Efficient Traffic Prediction Through Spatio-Temporal Distillation (AAAI 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.