Research map / Graph learning / GNN architectures
Dual-branch Spatial-Temporal Self-supervised Representation for Enhanced Road Network Learning
Qinghong Guo, Yu Wang, Ji Cao, Tongya Zheng, Junshu Dai, Bingde Hu, Shunyu Liu, Canghong Jin
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| use DST instead of TrajRNE for destination prediction | DST · Beijing destination prediction | ACC@1: 0.6728 -> 0.7288 | Improved |
| use DST instead of TrajRNE for destination prediction | DST · Porto destination prediction | ACC@1: 0.6728 -> 0.6766 | No clear effect |
| use DST instead of JCLRNT for speed inference | DST · Xian speed inference | MAE: 4.5138 -> 4.4987 | No clear effect |
| use DST instead of JCLRNT for speed inference | DST · Beijing speed inference | MAE: 2.8512 -> 2.4595 | Improved |
| add mix-hop transition matrix (Phop) (ablation) | DST · Beijing, speed inference | MAE/RMSE | Improved |
| add same-type hyperedges (hg2) (ablation) | DST · Beijing, speed inference | MAE/RMSE | Improved |
| add temporal traffic dynamics modeling (ablation) | DST · Beijing, destination prediction and travel time | ACC@1/MAE | Improved |
| add region and one-way hyperedges (hg1, hg3) (ablation) | DST · Beijing, all tasks | all metrics | Improved |
| transfer DST trained on Beijing to Porto without mix-hop matrix | DST-transfer · Porto zero-shot destination prediction | ACC@1: 0.0167 (JCLRNT) -> 0.6424 | Improved |
| increase next-token prediction loss weight in lambda ratio (ablation) | DST · Destination prediction | ACC@1/MRR | Improved |
Closest papers by meaning
- UniTR: A Unified Framework for Joint Representation Learning of Trajectories and Road Networks (AAAI 2025 · similarity 0.89)
- Towards Dynamic Spatial-Temporal Graph Learning: A Decoupled Perspective (AAAI 2024 · similarity 0.88)
- Road Network Representation Learning with the Third Law of Geography (NeurIPS 2024 · similarity 0.88)
- Efficient Traffic Prediction Through Spatio-Temporal Distillation (AAAI 2025 · similarity 0.88)
- Bridging Traffic State and Trajectory for Dynamic Road Network and Trajectory Representation Learning (AAAI 2025 · similarity 0.88)
- SSL-STMFormer Self-Supervised Learning Spatio-Temporal Entanglement Transformer for Traffic Flow Prediction (AAAI 2025 · similarity 0.87)
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.
