Research map / Label-efficient and robust learning / Knowledge distillation
Towards Learning Spatially Discriminative Feature Representations
Chaofei Wang, Jiayu Xiao, Yizeng Han, Qisen Yang, Shiji Song, Gao Huang
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 CAM-loss | ResNet-56 · CIFAR-100 | top-1 error (%): 28.80 -> 27.34 | Improved |
| add CAM-loss | ResNet-50 · ImageNet-1K | top-1 error (%): 23.68 -> 22.98 | Improved |
| add CAM-loss | ResNext-101 8x32d · ImageNet-1K | top-1 error (%): 21.04 -> 20.45 | Improved |
| add CAM-loss | DenseNet-bc-190-40 · CIFAR-100 | top-1 error (%): 17.67 -> 16.98 | Improved |
| add CAM-loss to ShakeDrop | PyramidNet-200 · CIFAR-100 | top-1 error (%): 15.08 -> 14.56 | Improved |
| add CAM-loss to CutMix | PyramidNet-200 · CIFAR-100 | top-1 error (%): 14.47 -> 14.01 | Improved |
| add CAM-loss to CutMix + ShakeDrop | PyramidNet-200 · CIFAR-100 | top-1 error (%): 13.81 -> 13.49 | Improved |
| add CAM-loss to CutMix | ResNet-50 · ImageNet-1K | top-1 error (%): 21.54 -> 21.16 | Improved |
| use CAM-loss instead of NegativeCAM | ResNet-50 · ImageNet-1K | top-1 error (%): 23.32 -> 22.98 | Improved |
| use CAM-loss instead of NegativeCAM | ResNet-56 · CIFAR-100 | top-1 error (%): 27.37 -> 27.34 | No clear effect |
| use CAM-loss instead of center loss | ResNet-110 · CIFAR-100 | top-1 error (%): 26.88 -> 26.56 | Improved |
| pretrain with CAM-loss then fine-tune | ResNet-50 · CUB transfer | top-1 accuracy (%): 85.6 -> 86.7 | Improved |
Closest papers by meaning
- Finer-CAM: Spotting the Difference Reveals Finer Details for Visual Explanation (CVPR 2025 · similarity 0.85)
- DD-CAM: Minimal Sufficient Explanations for Vision Models Using Delta Debugging (NeurIPS 2026 · similarity 0.85)
- Empowering CAM-Based Methods with Capability to Generate Fine-Grained and High-Faithfulness Explanations (AAAI 2024 · similarity 0.84)
- CAPE: CAM as a Probabilistic Ensemble for Enhanced DNN Interpretation (CVPR 2024 · similarity 0.84)
- Emphasizing Discriminative Features for Dataset Distillation in Complex Scenarios (CVPR 2025 · similarity 0.84)
- PICNN: A Pathway towards Interpretable Convolutional Neural Networks (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.
