Research map / Label-efficient and robust learning / Recognition and long-tail
Label-Aware Global Consistency for Multi-Label Learning with Single Positive Labels
Ming-Kun Xie, Jia-Hao Xiao, 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 pseudo-labeling consistency regularization PLC (ablation) | LEM with AN loss · VOC | mAP (%): 87.6 -> 89.0 | Improved |
| add pseudo-labeling consistency regularization PLC (ablation) | LEM with AN loss · COCO | mAP (%): 72.3 -> 73.7 | Improved |
| add LAC with end-to-end training (ablation) | LEM with AN loss · VOC | mAP (%): 89.0 -> 84.5 | Hurt |
| add LAC with end-to-end training (ablation) | LEM with AN loss · COCO | mAP (%): 73.7 -> 65.6 | Hurt |
| add LAC with two-stage training (ablation) | LEM with AN loss · VOC | mAP (%): 84.5 -> 89.2 | Improved |
| add LAC with two-stage training (ablation) | LEM with AN loss · COCO | mAP (%): 65.6 -> 74.7 | Improved |
| add memory queue (ablation) | LEM with AN loss · VOC | mAP (%): 89.2 -> 89.6 | Improved |
| add memory queue (ablation) | LEM with AN loss · COCO | mAP (%): 74.7 -> 75.6 | Improved |
| vary threshold delta, temperature tau and queue size K (ablation) | LEM with AN loss · COCO | mAP (%): 0.3% | No clear effect |
| vary temperature tau (ablation) | LEM with AN loss · NUS | mAP (%): 0.2% | No clear effect |
| vary memory queue size K (ablation) | LEM with AN loss · COCO | mAP (%): 0.2% | No clear effect |
| replace ResNet-50 by LEM for ROLE (ablation) | LEM with AN loss · VOC | mAP (%): 88.2 -> 88.1 | Hurt |
Closest papers by meaning
- Can Class-Priors Help Single-Positive Multi-Label Learning? (NeurIPS 2025 · similarity 0.88)
- Addressing Multi-Label Learning with Partial Labels: From Sample Selection to Label Selection (AAAI 2025 · similarity 0.88)
- Generative Modeling Reinvents Supervised Learning: Label Repurposing with Predictive Consistency Learning (ICML 2025 · similarity 0.88)
- Multi-Label Learning with Stronger Consistency Guarantees (NeurIPS 2024 · similarity 0.88)
- Controller-Guided Partial Label Consistency Regularization with Unlabeled Data (AAAI 2024 · similarity 0.87)
- Generalization Analysis for 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.
