Research map / Label-efficient and robust learning / Recognition and long-tail
One Coin Has Two Sides: Single Poistive Multi Label Learning from Salient Annotations
Xiaoyu Wang, Zhuoming Li, Bo Han, Hui LIU, Junhui Hou, Yuheng Jia
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 prototype-guided embedding loss (PL) (ablation) | PiSA · SalSPML confidence-based, avg of COCO/NUS/VOC | mAP: 59.49 -> 60.97 | Improved |
| add prototype-guided loss rejection (PGLR) (ablation) | PiSA · SalSPML confidence-based, avg of COCO/NUS/VOC | mAP: 59.49 -> 60.61 | Improved |
| add both PL and PGLR (ablation) | PiSA · SalSPML confidence-based, avg of COCO/NUS/VOC | mAP: 59.49 -> 62.63 | Improved |
| add PL and PGLR (ablation) | PiSA · SalSPML confidence-based VOC | mAP: 83.21 -> 87.12 | Improved |
| add PL and PGLR (ablation) | PiSA · SalSPML confidence-based COCO | mAP: 59.59 -> 63.43 | Improved |
| reject top-k unlabeled classes by prototype similarity instead of prediction probability (ablation) | PiSA · SalSPML real VOC (k=3) | mAP: +4.27% | Improved |
| reject top-k unlabeled classes by prototype similarity instead of prediction probability (ablation) | PiSA · SalSPML real COCO (k=3) | mAP: +0.79%, +0.97%, +2.09% | Improved |
| use adaptive threshold rejection (w=0.5, 1, 2) instead of fixed top-k=3 (ablation) | PiSA · SalSPML real COCO | mAP: 59.58 -> 59.21, 59.45, 58.97 | No clear effect |
| use adaptive threshold rejection (w=0.5, 1, 2) instead of fixed top-k=3 (ablation) | PiSA · SalSPML real VOC | mAP: 87.83 -> 87.32, 87.09, 85.53 | Hurt |
| apply PiSA instead of BCE | BCE baseline · SalSPML real COCO annotation bias | bias Delta_bias (lower better): 5.60 -> 4.57 | Improved |
| apply PiSA instead of BCE | BCE baseline · SalSPML real VOC annotation bias | bias Delta_bias (lower better): 3.40 -> 2.91 | Improved |
| apply PiSA instead of BCE | BCE baseline · SalSPML confidence-based COCO and VOC annotation bias | bias Delta_bias (lower better): 6.22 -> 4.49 (COCO), 4.10 -> 3.11 (VOC) | Improved |
Closest papers by meaning
- Acknowledging the Unknown for Multi-Label Learning with Single Positive Labels (ECCV 2022 · similarity 0.90)
- Multi-Label Learning From Single Positive Labels (CVPR 2021 · similarity 0.88)
- Revisiting Pseudo-Label for Single-Positive Multi-Label Learning (ICML 2023 · similarity 0.88)
- Addressing Multi-Label Learning with Partial Labels: From Sample Selection to Label Selection (AAAI 2025 · similarity 0.87)
- Learning with Partial-Label and Unlabeled Data: A Uniform Treatment for Supervision Redundancy and Insufficiency (ICML 2024 · similarity 0.86)
- Can Class-Priors Help Single-Positive Multi-Label Learning? (NeurIPS 2025 · 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.
