Research map / Learning theory and algorithms / Clustering
Correlation clustering
Develop approximation algorithms for correlation and hierarchical clustering on graphs.
102 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Clustering, keyword cluster “correlation clustering · hierarchical · spectral clustering”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
27 papers in 2023–24 and 23 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 19.3% to 11.2%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Clustering with Same-Cluster Queries (NeurIPS 2016 · 55 citations)
- 2017 · Inhomogeneous Hypergraph Clustering with Applications (NeurIPS 2017 · 94 citations)
- 2018 · Submodular Hypergraphs: p-Laplacians, Cheeger Inequalities and Spectral Clustering (ICML 2018 · 59 citations)
- 2019 · Efficient Parameter-Free Clustering Using First Neighbor Relations (CVPR 2019 · 255 citations)
- 2020 · Efficient Clustering Based On A Unified View Of $K$-means And Ratio-cut (NeurIPS 2020 · 14 citations)
- 2021 · Hierarchical Agglomerative Graph Clustering in Nearly-Linear Time (ICML 2021 · 10 citations)
- 2022 · RAMA: A Rapid Multicut Algorithm on GPU (CVPR 2022 · 7 citations)
- 2023 · Understanding the Generalization Performance of Spectral Clustering Algorithms (AAAI 2023 · 5 citations)
- 2024 · Equivalence between Graph Spectral Clustering and Column Subset Selection (Student Abstract) (AAAI 2024 · 2 citations)
- 2025 · Clustering via Hedonic Games: New Concepts and Algorithms (NeurIPS 2025 · 2 citations)
- 2026 · Sublinear Spectral Clustering Oracle with Little Memory (ICLR 2026 · 0 citations)
Earliest papers here
- Clustering with Same-Cluster Queries (NeurIPS 2016 · 55 citations)
- Clustering Signed Networks with the Geometric Mean of Laplacians (NeurIPS 2016 · 22 citations)
- Optimal Cluster Recovery in the Labeled Stochastic Block Model (NeurIPS 2016 · 14 citations)
What has been tried
86 papers here report 627 measured changes (a component added, removed or swapped, compared with a baseline): helped 69%, no clear effect 10%, hurt 16%, mixed 5%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Use ϵ=0 for the CSS-based clustering algorithm | Cockroach graph, k=2–6 · NCut | k=2: 0.069; k=3: 0.147; k=4: 0.322; k=5: 0.600; k=6: 0.878 | Mixed | Equivalence between Graph Spectral Clustering and Column Subset Selection (Student Abstract) (AAAI 2024 · 2 citations) |
| Add cluster-merging postprocessing to DegMFP | Netscience, Harvard500, football, Erdos991 · approximation ratio | 1.72 -> 1.58; 1.95 -> 1.79; 1.79 -> 1.47; 1.99 -> 1.67 | Helped | Combinatorial Approximations for Cluster Deletion: Simpler, Faster, and Better (ICML 2024 · 2 citations) |
| add the approximate 1NN heuristic | Mnist8m, L2 · NMI | 25% -> 36% | Helped | Scalable DBSCAN with Random Projections (NeurIPS 2024 · 2 citations) |
| Set epsilon to 0.2 instead of 0.1 | Benchmark datasets (Avg) · ARI | 0.471 -> 0.465 | No clear effect | Efficient Centroid-Linkage Clustering (NeurIPS 2024 · 1 citation) |
| Set α =0.1 instead of the default setting | 10 datasets (average) · NMI | 29.6 -> 28.0 | Hurt | Generalization Performance of Ensemble Clustering: From Theory to Algorithm (ICML 2025 · 0 citations) |
| Use a sweep cut of n/2 vertices instead of classifying by eigenvector signs | Benchmark distribution, sweep cut · agreement | 100% | No clear effect | On the Robustness of Spectral Algorithms for Semirandom Stochastic Block Models (NeurIPS 2024 · 0 citations) |
| Increase k-center component count from t=16 to t=256 | Cooking · runtime ratio | 4.391 -> 15.598 | Helped | Approximate Forest Completion and Learning-Augmented Algorithms for Metric Minimum Spanning Trees (ICML 2025) |
What reviewers will expect
Datasets and baselines named in the 16 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- DBLP · 2 of 16 papers
- mnist · 2 of 16 papers
- cifar10 · 2 of 16 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 50 papers; from OpenAlex and the papers' affiliations).
- Google · 7 papers
- Northwestern University · 4 papers
- Texas A&M · 4 papers
- University of Edinburgh · 3 papers
- Texas A&M University - College Station · 3 papers
- University of Science and Technology of China · 3 papers
- University of California, San Diego · 2 papers
- Pontificia Universidade Catolica · 2 papers
- University of Utah · 2 papers
- EPFL · 2 papers
Use it
Download BibTeX of 40 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Clustering approximation algorithms (150 papers)
- Subspace clustering (102 papers)
- Fairness-constrained clustering (19 papers)
- Incomplete multi-view clustering (125 papers)
- Spectral clustering algorithms (43 papers)
Elsewhere in the topic
- Coresets and dynamic clustering (77 papers)
- k-means++ seeding analysis (76 papers)
- Fair clustering approximation (31 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).