Research map / Label-efficient and robust learning / Recognition and long-tail
Supervised Exploratory Learning for Long-Tailed Visual Recognition
Zhongquan Jian, Yanhao Chen, Yancheng Wang, Junfeng Yao, Meihong Wang, Qingqiang Wu
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 Class-biased Complement (CbC) (ablation) | SEL on ResNet-32 · CIFAR-100-LT (imbalance 100) | top-1 accuracy: 39.96 -> 43.44 | Improved |
| add CbC (ablation) | SEL on ResNet-32 · CIFAR-100-LT (imbalance 10) | top-1 accuracy: 56.66 -> 57.84 | Improved |
| add Fitness-weighted Sampling (FwS) alone without CbC (ablation) | SEL on ResNet-32 · CIFAR-100-LT (imbalance 100) | top-1 accuracy: 39.96 -> 40.26 | No clear effect |
| add FwS on top of CbC (ablation) | SEL on ResNet-32 · CIFAR-100-LT (imbalance 100) | top-1 accuracy: 43.44 -> 44.53 | Improved |
| add FwS on top of CbC (ablation) | SEL on ResNet-32 · CIFAR-100-LT (imbalance 10) | top-1 accuracy: 57.84 -> 59.17 | Improved |
| add SEL instead of MixUp | ResNet-32 · CIFAR-100-LT (imbalance 100) | top-1 accuracy: 41.55 -> 44.53 | Improved |
| add SEL | GLMC · CIFAR-100-LT (imbalance 100) | top-1 accuracy: 53.91 -> 56.48 | Improved |
| add SEL | IB · CIFAR-10-LT (imbalance 100) | top-1 accuracy: 75.11 -> 75.04 | No clear effect |
| add SEL | H2T+CE · CIFAR-100-LT (imbalance 100) | top-1 accuracy: 42.27 -> 42.45 | No clear effect |
| add SEL | BCL · CIFAR-100-LT (imbalance 100) | top-1 accuracy: 51.76 -> 52.30 | Improved |
| add SEL | ResNet-50 · ImageNet-LT Few classes | top-1 accuracy: 26.04 -> 31.27 | Improved |
| add SEL | GLMC · ImageNet-LT Many classes | top-1 accuracy: 69.42 -> 68.67 | Hurt |
Closest papers by meaning
- Decoupled Optimisation for Long-Tailed Visual Recognition (AAAI 2024 · similarity 0.85)
- Feature Fusion from Head to Tail for Long-Tailed Visual Recognition (AAAI 2024 · similarity 0.85)
- Long-Tailed Learning as Multi-Objective Optimization (AAAI 2024 · similarity 0.84)
- ExPLoRA: Parameter-Efficient Extended Pre-Training to Adapt Vision Transformers under Domain Shifts (ICML 2025 · similarity 0.84)
- Learning from Neighbors: Category Extrapolation for Long-Tail Learning (CVPR 2025 · similarity 0.84)
- Long-tailed Recognition with Model Rebalancing (NeurIPS 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.
