Research map / Label-efficient and robust learning / Recognition and long-tail
Handling Long-tailed Feature Distribution in AdderNets
Minjing Dong, Yunhe Wang, Xinghao Chen, Chang Xu
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 SLAC loss (skew Laplace mixture with angle constraint) | VGG-small ANN · CIFAR-100 | accuracy: 72.64 -> 73.63 | Improved |
| add SLAC loss | ResNet-20 ANN · CIFAR-100 | accuracy: 67.60 -> 68.31 | Improved |
| add SLAC loss | ResNet-20 ANN · CIFAR-10 | accuracy: 91.84 -> 92.29 | Improved |
| add SLAC loss | ResNet-32 ANN · CIFAR-100 | accuracy: 69.02 -> 69.83 | Improved |
| add SLAC loss | ResNet-18 ANN · ImageNet | top-1 accuracy: 67.0 -> 67.7 | Improved |
| add SLAC loss | ResNet-50 ANN · ImageNet | top-5 accuracy: 91.7 -> 92.6 | Improved |
| replace L-SoftMax margin loss with SLAC (ablation) | ResNet-20 ANN · CIFAR-100 | accuracy: 67.53 -> 68.31 | Improved |
| replace single Gaussian mixture (LGM) with SLAC (ablation) | ResNet-20 ANN · CIFAR-10 | accuracy: 92.00 -> 92.29 | Improved |
| enlarge distribution space with mixture of Gaussian mixtures (MoM) (ablation) | ResNet-20 ANN · CIFAR-100 | accuracy: 67.60 -> 67.73 | No clear effect |
| add angle-based constraint to skew Laplace distribution (ablation) | ResNet-20 ANN · CIFAR-100 | accuracy: 67.51 -> 68.31 | Improved |
| fix skewness during training (ablation) | SLAC-ANN · CIFAR-100 | accuracy: 68.31 -> 62.04 | Hurt |
| fix location during training (ablation) | SLAC-ANN · CIFAR-100 | accuracy: 68.31 -> 64.82 | Hurt |
Closest papers by meaning
- Feature Fusion from Head to Tail for Long-Tailed Visual Recognition (AAAI 2024 · similarity 0.86)
- No One Left Behind: Improving the Worst Categories in Long-Tailed Learning (CVPR 2023 · similarity 0.85)
- Learning from Neighbors: Category Extrapolation for Long-Tail Learning (CVPR 2025 · similarity 0.85)
- DiffuLT: Diffusion for Long-tail Recognition Without External Knowledge (NeurIPS 2024 · similarity 0.84)
- Towards Scalable and Versatile Weight Space Learning (ICML 2024 · similarity 0.83)
- Long-Tailed Learning as Multi-Objective Optimization (AAAI 2024 · similarity 0.83)
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.
