Research map / Label-efficient and robust learning / Recognition and long-tail
Calibratable Disambiguation Loss for Multi-Instance Partial-Label Learning
Wei Tang, Yi Yang, Weijia Zhang, Min-Ling Zhang
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 CDL to MIPL methods | DEMIPL, ELIMIPL, MIPLMA · benchmark datasets, r=1..3 | accuracy: outperform in 67 of 72 cases | Improved |
| add CDL | MIPL methods · benchmark datasets | accuracy: 3.27%, 4.57%, 8.93% mean gain for r=1,2,3 | Improved |
| add CDL-CC to DEMIPL | DEMIPL · MNIST-MIPL r=3 | accuracy: 70.93 -> 85.27 | Improved |
| add CDL-CN to MIPLMA | MIPLMA · MNIST-MIPL r=3 | accuracy: 74.93 -> 98.00 | Improved |
| add CDL-CN to DEMIPL | DEMIPL · SIVAL-MIPL r=3 | accuracy: 50.31 -> 68.89 | Improved |
| add CDL-CN to DEMIPL | DEMIPL · C-KMeans | accuracy: 52.11 -> 64.78 | Improved |
| add CDL-CC to ELIMIPL | ELIMIPL · FMNIST-MIPL r=1 | accuracy: 90.27 -> 90.27 | No clear effect |
| add CDL-CC to DEMIPL | DEMIPL · MNIST-MIPL r=1 | ECE (%): 59.40 -> 1.08 | Improved |
| add CDL-CN to MIPLMA | MIPLMA · FMNIST-MIPL r=1 | ECE (%): 52.38 -> 4.18 | Improved |
| add CDL-CN to MIPLMA | MIPLMA · SIVAL-MIPL r=1 | ECE (%): 61.67 -> 15.91 | Improved |
| compare FastMIPL to CDL variants | FastMIPL · benchmark datasets | accuracy: FastMIPL better in 4 of 72 cases | Hurt |
| add CDL-CC to DEMIPL | DEMIPL · C-SIFT | accuracy: 53.17 -> 53.59 | No clear effect |
Closest papers by meaning
- Calibrated Disambiguation for Partial Multi-label Learning (AAAI 2025 · similarity 0.90)
- Multi-Instance Partial-Label Learning with Margin Adjustment (NeurIPS 2024 · similarity 0.89)
- Fast Multi-Instance Partial-Label Learning (AAAI 2025 · similarity 0.89)
- Distilling Reliable Knowledge for Instance-Dependent Partial Label Learning (AAAI 2024 · similarity 0.89)
- Adaptive Integration of Partial Label Learning and Negative Learning for Enhanced Noisy Label Learning (AAAI 2024 · similarity 0.86)
- Controller-Guided Partial Label Consistency Regularization with Unlabeled Data (AAAI 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.
