Research map / Label-efficient and robust learning / Recognition and long-tail
CHB: A Diagnostic Toolkit for Hardness-Aware Clustering Evaluation
Walid Durani, Philipp Jahn, Collin Leiber, David B. Hoffmann, Thomas Seidl, Claudia Plant, Christian Böhm
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 |
|---|---|---|---|
| move representation from raw to best learned (UMAP/DDC) for density methods on CIFAR-10 | density clustering · CIFAR-10 | NMI: 0.033 -> 0.176 | Improved |
| move representation from raw to best learned (UMAP/DDC) for density methods on CIFAR-100 | density clustering · CIFAR-100 | NMI: 0.016 -> 0.351 | Improved |
| add CLIP then UMAP representation pipeline | clustering on CIFAR-10 · CIFAR-10 | Delta NMI: +.713 | Improved |
| add CLIP then UMAP representation pipeline | clustering on SVHN · SVHN | Delta NMI: -.044 | No clear effect |
| separability regime improves from A to B to C | all clustering methods · CHB regimes | median agreement: 0.067 -> 0.529 -> 0.704 | Improved |
| use deep-pipeline (DDC) instead of Spectral in Regime B | clustering family choice · Regime B | median agreement: 0.631 -> 0.695 | Improved |
| use Spectral instead of deep-pipeline in Regime C | clustering family choice · Regime C | median agreement: 0.727 -> 0.782 | Improved |
| use density-based methods in Regime B | clustering family choice · Regime B | median agreement: 0.05 | Hurt |
| apply mismatch-correcting embeddings without separability rescue (CIFAR/STL/SVHN) | clustering on Regime A datasets · CIFAR-10/100, STL-10, SVHN | NMI | Mixed |
Closest papers by meaning
- Dissecting Sample Hardness: A Fine-Grained Analysis of Hardness Characterization Methods for Data-Centric AI (ICLR 2024 · similarity 0.88)
- Stable and Consistent Density-Based Clustering via Multiparameter Persistence (JMLR 2024 · similarity 0.85)
- You Can Trust Your Clustering Model: A Parameter-free Self-Boosting Plug-in for Deep Clustering (NeurIPS 2025 · similarity 0.84)
- Scalable DBSCAN with Random Projections (NeurIPS 2024 · similarity 0.84)
- Structure-Aware Spectral Sparsification via Uniform Edge Sampling (NeurIPS 2025 · similarity 0.84)
- THESAURUS: Contrastive Graph Clustering by Swapping Fused Gromov-Wasserstein Couplings (AAAI 2025 · similarity 0.84)
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.
