Research map / Label-efficient and robust learning / Recognition and long-tail
Learning Bottleneck Concepts in Image Classification
Bowen Wang, Liangzhi Li, Yuta Nakashima, Hajime Nagahara
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 |
|---|---|---|---|
| Use contrastive rather than reconstruction loss for concept self-supervision (ablation) | BotCL · CUB200 | accuracy: 0.693 -> 0.740 | Improved |
| Use contrastive rather than reconstruction loss for concept self-supervision (ablation) | BotCL · ImageNet (first 200 classes) | accuracy: 0.720 -> 0.795 | Improved |
| Use reconstruction rather than contrastive loss for concept self-supervision (ablation) | BotCL · MNIST | accuracy: 0.980 -> 0.983 | Improved |
| Use contrastive rather than reconstruction loss for concept self-supervision (ablation) | BotCL · Synthetic | accuracy: 0.785 -> 0.998 | Improved |
| Increase the number of concepts (ablation) | BotCL · Synthetic | accuracy | Improved |
| Increase the quantization-loss weight (ablation) | BotCL · CUB200, ImageNet, and Synthetic | accuracy | Hurt |
| Add individual-consistency and mutual-distinctiveness regularizers (ablation) | BotCL · CUB200 and ImageNet | accuracy | No clear effect |
| Increase individual-consistency and mutual-distinctiveness loss weights substantially (ablation) | BotCL · All evaluated datasets | accuracy | Hurt |
| Use a larger number of concepts (ablation) | BotCL · All datasets | accuracy | Mixed |
Closest papers by meaning
- Show and Tell: Visually Explainable Deep Neural Nets via Spatially-Aware Concept Bottleneck Models (CVPR 2025 · similarity 0.91)
- Coarse-to-Fine Concept Bottleneck Models (NeurIPS 2024 · similarity 0.89)
- Explanation Bottleneck Models (AAAI 2025 · similarity 0.87)
- Interpretable Generative Models through Post-hoc Concept Bottlenecks (CVPR 2025 · similarity 0.87)
- Enhancing Performance of Explainable AI Models with Constrained Concept Refinement (ICML 2025 · similarity 0.87)
- Interpretable Concept-Based Memory Reasoning (NeurIPS 2024 · similarity 0.87)
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.
