Research map / Label-efficient and robust learning / Recognition and long-tail
Comprehensive Urban Region Representation Learning via Multi-View Joint Learning and Contrastive Learning
Yingde Lin, Yuanbo Xu, Lu Jiang, Pengyang Wang
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 |
|---|---|---|---|
| add the multi-view joint learning module (ablation) | MVJC · Crime prediction (NYC) | R2: 0.33 -> 0.70 | Improved |
| add the structure-aware contrastive learning module (ablation) | MVJC · Land use clustering (NYC) | ARI: ARI decreases by 45.8% | Improved |
| add the joint learning module (clustering effect) (ablation) | MVJC · Land use clustering (NYC) | clustering metrics | No clear effect |
| add the mobility view (ablation) | MVJC · Crime prediction (NYC) | MAE: 96.75 -> 60.80 | Improved |
| add the POI view (ablation) | MVJC · Crime prediction (NYC) | MAE: 78.19 -> 60.80 | Improved |
| add the check-in view (ablation) | MVJC · Crime prediction (NYC) | MAE: 64.35 -> 60.80 | Improved |
| add the check-in view (ablation) | MVJC · Land use clustering (NYC) | F-measure: 0.65 -> 0.70 | Improved |
| add the POI view (ablation) | MVJC · Land use clustering (NYC) | F-measure: 0.59 -> 0.70 | Improved |
| add the mobility view (ablation) | MVJC · Crime prediction (NYC) | R2: 0.42 -> 0.70 | Improved |
| use MVJC instead of ReCP | MVJC · Crime prediction (NYC) | MAE: 66.90 -> 60.80 (9.12%) | Improved |
| use MVJC instead of ReCP | MVJC · Land use clustering (NYC) | F-measure: 0.45 -> 0.70 (55.6%) | Improved |
| use w/o-Chk variant instead of best baseline | MVJC without check-ins · Land use clustering (NYC) | F-measure: improving clustering F-measure by 44.44% over the best baseline | Improved |
Closest papers by meaning
- Urban Region Embedding via Multi-View Contrastive Prediction (AAAI 2024 · similarity 0.91)
- UniTR: A Unified Framework for Joint Representation Learning of Trajectories and Road Networks (AAAI 2025 · similarity 0.85)
- Improving Region Representation Learning from Urban Imagery with Noisy Long-Caption Supervision (AAAI 2026 · similarity 0.84)
- Road Network Representation Learning with the Third Law of Geography (NeurIPS 2024 · similarity 0.84)
- Cross-City Latent Space Alignment for Consistency Region Embedding (ICML 2025 · similarity 0.84)
- UrBench: A Comprehensive Benchmark for Evaluating Large Multimodal Models in Multi-View Urban Scenarios (AAAI 2025 · similarity 0.83)
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.
