Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Keep Your Friends Close, and the Right Neighbours Closer: Disaster-Conditioned Kernel-Regularized Graph Attention for Building Damage Classification
Fuad Hasan, Chul Min Yeum
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 graph attention over neighbouring buildings (vanilla GAT) | patch-only encoder · xBD LOEO | Macro-F1: 0.433 -> 0.453 | Improved |
| add graph attention (vanilla GAT) | patch-only encoder · xBD LOEO | Residual Moran's I: 0.248 -> 0.256 | No clear effect |
| add DCMS kernel prior (ablation) | vanilla GAT · xBD LOEO | Macro-F1: 0.453 -> 0.473 | Improved |
| add DCMS kernel prior (ablation) | vanilla GAT · xBD LOEO | Residual Moran's I: 0.256 -> 0.182 | Improved |
| add disaster-type conditioning (ablation) | vanilla GAT · xBD LOEO | Macro-F1: 0.453 -> 0.460 | Improved |
| add residual Moran de-correlation loss (ablation) | vanilla GAT · xBD LOEO | Residual Moran's I: 0.256 -> 0.139 | Improved |
| add residual Moran de-correlation loss (ablation) | vanilla GAT · xBD LOEO | Macro-F1: 0.453 -> 0.450 | No clear effect |
| add disaster conditioning on top of kernel + residual (ablation) | kernel + residual model · xBD LOEO | Macro-F1: 0.490 -> 0.503 | Improved |
| add residual loss on top of kernel + disaster conditioning (ablation) | kernel + disaster model · xBD LOEO | Residual Moran's I: 0.116 -> 0.079 | Improved |
| use pre+post bitemporal input instead of post-only (ablation) | Ours · xView2 holdout (fixed instances) | Macro-F1: 0.828 -> 0.873 | Improved |
| vary kNN k from 8 to 16 to 32 (ablation) | Ours · xBD LOEO | Macro-F1: 0.498, 0.503, 0.500 | No clear effect |
| apply Ours vs patch-only to unseen dataset | Ours · xBD to Ida-BD | Macro-F1: 0.275 -> 0.335 | Improved |
Closest papers by meaning
- Generalizable Disaster Damage Assessment via Change Detection with Vision Foundation Model (AAAI 2025 · similarity 0.85)
- FireScope: Wildfire Risk Raster Prediction With a Chain-of-Thought Oracle (CVPR 2026 · similarity 0.84)
- DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response (NeurIPS 2025 · similarity 0.84)
- Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations (NeurIPS 2026 · similarity 0.84)
- DISCount: Counting in Large Image Collections with Detector-Based Importance Sampling (AAAI 2024 · similarity 0.84)
- Graph Structure Learning for Spatial-Temporal Imputation: Adapting to Node and Feature Scales (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.
