Research map / Label-efficient and robust learning / Recognition and long-tail
Decompositional Generation Process for Instance-Dependent Partial Label Learning
Congyu Qiao, Ning Xu, Xin Geng
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 |
|---|---|---|---|
| model candidate-label generation as instance-dependent categorical correct-label selection followed by Bernoulli incorrect-label selection, with MAP priors | PLL classification · MNIST | accuracy: 98.84±0.05% -> 98.92±0.05% | Improved |
| model candidate-label generation as instance-dependent categorical correct-label selection followed by Bernoulli incorrect-label selection, with MAP priors | PLL classification · FMNIST | accuracy: 90.63±0.30% -> 91.48±0.32% | Improved |
| model candidate-label generation as instance-dependent categorical correct-label selection followed by Bernoulli incorrect-label selection, with MAP priors | PLL classification · KMNIST | accuracy: 96.19±0.75% -> 96.88±0.17% | Improved |
| model candidate-label generation as instance-dependent categorical correct-label selection followed by Bernoulli incorrect-label selection, with MAP priors | PLL classification · CIFAR-10 | accuracy: 86.37±0.38% -> 87.57±0.26% | Improved |
| model candidate-label generation as instance-dependent categorical correct-label selection followed by Bernoulli incorrect-label selection, with MAP priors | PLL classification · CIFAR-100 | accuracy: 64.12±0.23% -> 64.59±0.17% | Improved |
| use instance-dependent decompositional generation and MAP priors | PLL classification · Lost | accuracy: 76.87±0.86% -> 77.02±0.82% | No clear effect |
| use instance-dependent decompositional generation and MAP priors | PLL classification · BirdSong | accuracy: 73.47±0.13% -> 74.23±0.17% | Improved |
| use instance-dependent decompositional generation and MAP priors | PLL classification · MSRCv2 | accuracy: 49.97±0.43% -> 50.45±0.47% | Improved |
| use instance-dependent decompositional generation and MAP priors | PLL classification · Soccer Player | accuracy: 56.02±0.59% -> 55.99±0.28% | Hurt |
| use instance-dependent decompositional generation and MAP priors | PLL classification · Yahoo!News | accuracy: 66.26±0.13% -> 66.62±0.19% | Improved |
| add prior regularization and iteratively refine prior information (ablation) | PL-ML · Lost | accuracy: 52.50±3.20% -> 77.02±0.82% | Improved |
| add prior regularization and iteratively refine prior information (ablation) | PL-ML · BirdSong | accuracy: 71.13±0.45% -> 74.23±0.17% | Improved |
Closest papers by meaning
- Reduction-based Pseudo-label Generation for Instance-dependent Partial Label Learning (NeurIPS 2025 · similarity 0.92)
- Realistic Evaluation of Deep Partial-Label Learning Algorithms (ICLR 2025 · similarity 0.90)
- Conformal Prediction for Partial Label Learning (AAAI 2025 · similarity 0.89)
- Partial Label Causal Representation Learning for Instance-Dependent Supervision and Domain Generalization (AAAI 2025 · similarity 0.89)
- Distilling Reliable Knowledge for Instance-Dependent Partial Label Learning (AAAI 2024 · similarity 0.89)
- What Makes Partial-Label Learning Algorithms Effective? (NeurIPS 2024 · similarity 0.89)
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.
