Research map / Label-efficient and robust learning / Knowledge distillation
Multi-Label Knowledge Distillation
Penghui Yang, Ming-Kun Xie, Chen-Chen Zong, Lei Feng, Gang Niu, Masashi Sugiyama, Sheng-Jun 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 multi-label logits distillation (MLD) to the student (ablation) | ResNet-34 student · MS-COCO | mAP: 70.31 -> 70.68 | Improved |
| add class-aware label-wise embedding distillation (CD) on top of MLD (ablation) | L2D · MS-COCO | mAP: 70.68 -> 71.91 | Improved |
| add instance-aware label-wise embedding distillation (ID) on top of MLD (ablation) | L2D · MS-COCO | mAP: 70.68 -> 71.79 | Improved |
| combine MLD, CD and ID (ablation) | L2D · MS-COCO | mAP: 71.91 -> 72.87 | Improved |
| use conventional feature-based KD (RKD, PKT, ReviewKD) for multi-label | ResNet-34 student · MS-COCO | mAP: 70.31 -> 70.39 (ReviewKD) | No clear effect |
| use L2D instead of the best baseline PS on different architectures | Swin-T teacher, MobileNet v2 student · MS-COCO | mAP: 72.42 -> 74.21 | Improved |
| distill with L2D on same-architecture WRN pair | WRN-50 student · MS-COCO | mAP: 74.45 -> 76.61 | Improved |
| apply L2D when the vanilla student outperforms the teacher (reversed KD) | ResNet-101 student · MS-COCO | mAP: 73.98 -> 75.51 | Improved |
| apply L2D on VOC with fewer labels per image | ResNet-18 student · VOC2007 | mAP: 84.01 -> 85.71 | Improved |
| L2D balancing parameters varied | L2D · MS-COCO | mAP | No clear effect |
Closest papers by meaning
- Multi-label Self Knowledge Distillation (AAAI 2025 · similarity 0.89)
- Distilling Reliable Knowledge for Instance-Dependent Partial Label Learning (AAAI 2024 · similarity 0.88)
- Rethinking Self-Distillation: Label Averaging and Enhanced Soft Label Refinement with Partial Labels (ICLR 2025 · similarity 0.87)
- Predicting Label Distribution from Ternary Labels (NeurIPS 2024 · similarity 0.86)
- Specifying What You Know or Not for Multi-Label Class-Incremental Learning (AAAI 2025 · similarity 0.86)
- Scaled Decoupled Distillation (CVPR 2024 · similarity 0.86)
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.
