Research map / Label-efficient and robust learning / Recognition and long-tail
Long-Tailed Classification with Multi-Granularity Semantics
Yuting Liu, Liu Yang, Yu Wang
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 LLM-label-semantic multi-granularity contrastive learning versus ConCutMix | SKCL (ResNet-32) · CIFAR-100-LT (imbalance 100) | Top-1 accuracy (%): 53.16 -> 54.02 | Improved |
| add SKCL on top of the BCL baseline | ResNet-50 · ImageNet-LT | Few-shot accuracy (%): 36.5 -> 41.5 | Improved |
| add SKCL improves head classes as well as tail | SKCL (ResNet-32) · CIFAR-100-LT (imbalance 100) | Many-shot accuracy (%): 67.4 -> 68.3 | Improved |
| add coarse-to-fine (vertical) transfer only (ablation) | SKCL · CIFAR-100-LT (imbalance 100) | overall accuracy (%): 51.9 -> 53.1 | Improved |
| add same-level (horizontal) transfer only (ablation) | SKCL · CIFAR-100-LT (imbalance 100) | overall accuracy (%): 51.9 -> 53.2 | Improved |
| combine vertical and horizontal transfer (ablation) | SKCL · CIFAR-100-LT (imbalance 100) | overall accuracy (%): 53.2 -> 54.0 | Improved |
| use LLM-generated label descriptions instead of original label semantics for similarity | SKCL · CIFAR-10-LT | similarity matrix (qualitative) | Improved |
| restrict transfer to Top-K similar classes | SKCL · long-tailed recognition | training behavior (qualitative) | Improved |
Closest papers by meaning
- Learning from Neighbors: Category Extrapolation for Long-Tail Learning (CVPR 2025 · similarity 0.87)
- Continuous Contrastive Learning for Long-Tailed Semi-Supervised Recognition (NeurIPS 2024 · similarity 0.86)
- Decoupled Contrastive Learning for Long-Tailed Recognition (AAAI 2024 · similarity 0.85)
- LTGC: Long-tail Recognition via Leveraging LLMs-driven Generated Content (CVPR 2024 · similarity 0.84)
- Feature Fusion from Head to Tail for Long-Tailed Visual Recognition (AAAI 2024 · similarity 0.84)
- Weakly-Supervised Contrastive Learning for Imprecise Class Labels (ICML 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.
