Research map / Label-efficient and robust learning / Recognition and long-tail
Towards Reliable Multi-Label Classification via Conditional Dependency Modeling
Arkapal Panda, Aditya Shankar Pal, Utpal Garain
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 the PCD auxiliary loss to BCE (CMLL) | BCE (ResNet-50) · PASCAL VOC 2012 | ACE: 0.1533 -> 0.1265 | Improved |
| add the PCD auxiliary loss to BCE (CMLL) | BCE (ViT-B/32) · PASCAL VOC 2012 | ACE: 0.0824 -> 0.0247 | Improved |
| add the PCD auxiliary loss to BCE (CMLL) | BCE (ResNet-50) · MS-COCO | ACE: 0.0654 -> 0.0030 | Improved |
| add the PCD auxiliary loss to BCE (CMLL) | BCE (ViT-B/32) · MS-COCO | ACE: 0.0829 -> 0.0020 | Improved |
| add the PCD auxiliary loss to BCE (CMLL) | BCE (ResNet-50) · WIDER-A | ACE: 0.1352 -> 0.1232 | Improved |
| add the PCD auxiliary loss to BCE (CMLL) | BCE (ViT-B/32) · WIDER-A | ACE: 0.0795 -> 0.0145 | Improved |
| add the PCD auxiliary loss to BCE (CMLL) | BCE (ViT-B/32) · PASCAL VOC 2012 | MCE: 0.2346 -> 0.0761 | Improved |
| add the PCD auxiliary loss to BCE (CMLL) | BCE (ResNet-50) · MS-COCO | MCE: 0.1766 -> 0.0052 | Improved |
| add the PCD auxiliary loss to BCE (CMLL) | BCE (ViT-B/32) · MS-COCO | mAP: 0.9207 -> 0.9775 | Improved |
| add the PCD auxiliary loss to BCE (CMLL) | BCE (ResNet-50) · PASCAL VOC 2012 | mAP: 0.9025 -> 0.9251 | No clear effect |
| add the PCD auxiliary loss to BCE (CMLL) | BCE (ResNet-50) · WIDER-A | mAP: 0.8198 -> 0.8248 | No clear effect |
| vary the number of bins used to compute ACE and MCE | CMLL · PASCAL VOC 2012, MS-COCO, WIDER-A | ACE/MCE | No clear effect |
Closest papers by meaning
- Holistic Label Correction for Noisy Multi-Label Classification (ICCV 2023 · similarity 0.92)
- Multi-Label Learning with Stronger Consistency Guarantees (NeurIPS 2024 · similarity 0.89)
- Predicting Label Distribution from Ternary Labels (NeurIPS 2024 · similarity 0.89)
- Addressing Multi-Label Learning with Partial Labels: From Sample Selection to Label Selection (AAAI 2025 · similarity 0.89)
- Limited-Supervised Multi-Label Learning with Dependency Noise (AAAI 2024 · similarity 0.88)
- Correlation-Induced Label Prior for Semi-Supervised Multi-Label Learning (ICML 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.
