Research map / Learning theory and algorithms / Clustering
GASP, a Generalized Framework for Agglomerative Clustering of Signed Graphs and Its Application to Instance Segmentation
Check 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 |
|---|---|---|---|
| Use average linkage instead of sum linkage on CNN-predicted grid graphs | GASP for neuron instance segmentation · CREMI-gridGraph (OurCNN) | ARAND Error: 0.0856 -> 0.0487 | Helped |
| Use average linkage instead of sum linkage on superpixel region-adjacency graphs | GASP for neuron instance segmentation · CREMI-3D-RAG (OurCNN) | ARAND Error: 0.0905 -> 0.0896 | Helped |
| Use average linkage instead of AbsMax linkage | GASP for neuron instance segmentation · CREMI-gridGraph (OurCNN) | ARAND Error: 0.0554 -> 0.0487 | Helped |
| Use average linkage instead of single linkage | GASP for neuron instance segmentation · CREMI-gridGraph (OurCNN) | ARAND Error: 0.9264 -> 0.0487 | Helped |
| Use average linkage instead of complete linkage | GASP for neuron instance segmentation · CREMI-3D-RAG (OurCNN) | ARAND Error: 0.5654 -> 0.0896 | Helped |
| Run average-linkage clustering directly on pixels rather than on superpixel 3D-RAGs | OurCNN + GASP · CREMI Challenge leaderboard | CREMI Score: 0.257 -> 0.224 | Helped |
| Use logarithmic mapping rather than mapping affinities as w_e=a_e-0.5 (ablation) | CNN edge-affinity conversion · CREMI-gridGraph | ARAND Error | Hurt |
| Add cannot-link constraints to average-linkage clustering (ablation) | HC-Avg · CREMI-gridGraph (OurCNN) | ARAND Error: 0.0487 -> 0.0492 | Hurt |
| Add cannot-link constraints to average-linkage clustering (ablation) | HC-Avg · CREMI-3D-RAG (OurCNN) | ARAND Error: 0.0896 -> 0.0898 | Hurt |
| Use sum linkage rather than average linkage on CNN-predicted grid graphs | GASP for neuron instance segmentation · CREMI-gridGraph (OurCNN) | ARAND Error: 0.0487 -> 0.0856 | Hurt |
Closest papers by meaning
- Hierarchical Agglomerative Graph Clustering in Nearly-Linear Time (ICML 2021 · similarity 0.89)
- Hierarchical Agglomerative Graph Clustering in Poly-Logarithmic Depth (NeurIPS 2022 · similarity 0.89)
- Optimal Graph Clustering without Edge Density Signals (NeurIPS 2025 · similarity 0.86)
- Dynamic Correlation Clustering in Sublinear Update Time (ICML 2024 · similarity 0.86)
- Hierarchical Overlapping Clustering on Graphs: Cost Function, Algorithm and Scalability (ICML 2025 · similarity 0.86)
- Edge-Colored Clustering in Hypergraphs: Beyond Minimizing Unsatisfied Edges (ICML 2025 · similarity 0.86)
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.