Research map / Label-efficient and robust learning / Semi-supervised learning
Discover and Align Taxonomic Context Priors for Open-world Semi-Supervised Learning
Yu Wang, Zhun Zhong, Pengchong Qiao, Xuxin Cheng, Xiawu Zheng, Chang Liu, Nicu Sebe, Rongrong Ji, Jie Chen
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 coarse-grained clustering without alignment (TCA) (ablation) | TIDA · CIFAR100 | All accuracy: 57.9 -> 55.5 | Hurt |
| add coarse-grained clustering with TCA (ablation) | TIDA · CIFAR100 | All accuracy: 57.9 -> 61.6 | Improved |
| add fine-grained clustering without TCA (ablation) | TIDA · CIFAR100 | All accuracy: 57.9 -> 55.3 | Hurt |
| add fine-grained clustering with TCA (ablation) | TIDA · CIFAR100 | All accuracy: 57.9 -> 63.1 | Improved |
| use both C-TCP and F-TCP with TCA (full model) (ablation) | TIDA · CIFAR100 | All accuracy: 63.1 -> 65.3 | Improved |
| use both C-TCP and F-TCP without TCA (ablation) | TIDA · CIFAR100 | All accuracy: 57.9 -> 57.7 | Hurt |
| add TCA to fine-grained clustering (ablation) | TIDA · Tiny ImageNet | All accuracy: 28.8 -> 36.8 | Improved |
| add TCA to coarse-grained clustering (ablation) | TIDA · Stanford-Cars | All accuracy: 61.1 -> 62.9 | Improved |
| three target-grained clusters aligned instead of multi-granularity (ablation) | TIDA · CIFAR100 | All accuracy: 65.3 -> 60.5 | Hurt |
| three target-grained clusters aligned instead of multi-granularity (ablation) | TIDA · Stanford-Cars | All accuracy: 66.0 -> 61.9 | Hurt |
| TIDA vs TRSSL | TRSSL · Tiny ImageNet | All accuracy: 30.3 -> 37.2 | Improved |
| TIDA vs TRSSL | TRSSL · Stanford-Cars | All accuracy: 60.4 -> 66.0 | Improved |
Closest papers by meaning
- Targeted Representation Alignment for Open-World Semi-Supervised Learning (CVPR 2024 · similarity 0.86)
- OwMatch: Conditional Self-Labeling with Consistency for Open-World Semi-Supervised Learning (NeurIPS 2024 · similarity 0.85)
- Bidirectional Logits Tree: Pursuing Granularity Reconcilement in Fine-Grained Classification (AAAI 2025 · similarity 0.85)
- CGMatch: A Different Perspective of Semi-supervised Learning (CVPR 2025 · similarity 0.85)
- Deep Taxonomic Networks for Unsupervised Hierarchical Prototype Discovery (NeurIPS 2025 · similarity 0.84)
- Self-Paced Unified Representation Learning for Hierarchical Multi-Label Classification (AAAI 2024 · 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.
