Research map / Label-efficient and robust learning / Recognition and long-tail
Topology-Aware Contrastive Learning: Regulating Representation Connectivity via Persistent Homology
Jiaxin Sun, Yuhua Qian, Yang 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 topology-aware connectivity regulation | SimCLR · ImageNet-100 | k-NN purity (k=3): 30.59 -> 37.19 | Improved |
| add topology-aware connectivity regulation | SimCLR · ImageNet-100 | linear accuracy: 60.98 -> 64.14 | Improved |
| add topology-aware connectivity regulation | SimCLR · CIFAR-100 | k-NN purity (k=3): 57.73 -> 60.36 | Improved |
| add topology-aware connectivity regulation | SimCLR · CIFAR-100 | linear accuracy: 60.38 -> 62.47 | Improved |
| add topology-aware connectivity regulation | SimCLR · CIFAR-10 | k-NN purity (k=3): 71.24 -> 71.77 | Improved |
| add topology-aware connectivity regulation | SimCLR · STL-10 | k-NN purity (k=3): 74.59 -> 75.27 | Improved |
| add topology-aware connectivity regulation under weak augmentation | SimCLR · CIFAR-100 weak (a)+(c) | Top-1 accuracy: 38.92 -> 43.19 | Improved |
| add topology-aware connectivity regulation under strong augmentation | SimCLR · CIFAR-100 strong (b)+(d) | Top-1 accuracy: 53.70 -> 55.72 | Improved |
| add topology-aware connectivity regulation under strong color jitter | SimCLR · CIFAR-100 strong (d) | Top-1 accuracy: 11.33 -> 17.35 | Improved |
| add topology-aware connectivity regulation under weak crop | SimCLR · CIFAR-100 weak (a) | Top-1 accuracy: 32.98 -> 35.13 | Improved |
| add topology-aware connectivity regulation on Vehicles 1 superclass with k=20 | SimCLR · CIFAR-100 Vehicles 1 | k-NN purity (k=20): 59.07 -> 60.49 | Improved |
| add topology-aware connectivity regulation on Fish superclass with k=8 | SimCLR · CIFAR-100 Fish | k-NN purity (k=8): 61.25 -> 62.40 | Improved |
Closest papers by meaning
- HTG-GCL: Leveraging Hierarchical Topological Granularity from Cellular Complexes for Graph Contrastive Learning (AAAI 2026 · similarity 0.89)
- Connectivity-Optimized Representation Learning via Persistent Homology (ICML 2019 · similarity 0.89)
- TopoGCL: Topological Graph Contrastive Learning (AAAI 2024 · similarity 0.87)
- Topological Blindspots: Understanding and Extending Topological Deep Learning Through the Lens of Expressivity (ICLR 2025 · similarity 0.87)
- Hierarchical Topology Isomorphism Expertise Embedded Graph Contrastive Learning (AAAI 2024 · similarity 0.87)
- Graphcode: Learning from multiparameter persistent homology using graph neural networks (NeurIPS 2024 · similarity 0.85)
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.
