Research map / Label-efficient and robust learning / Recognition and long-tail
Evidential Deep Partial Label Learning to Quantify Disambiguation Uncertainty
Jinfu Fan, Jiangnan Li, Xiaohui Zhong, Kangrui Ren, Zhencun Jiang, Min Gan, Tianhao Gu, Linqing 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 |
|---|---|---|---|
| replace PiCO loss with ED-PLL loss (PiCO-ED) | PiCO · CIFAR-10 (q=0.3) | disambiguation accuracy: 92.29 -> 92.90 | Improved |
| replace DIRK loss with ED-PLL loss (DIRK-ED) | DIRK · K-MNIST (q=0.3) | disambiguation accuracy: 96.49 -> 98.79 | Improved |
| add non-candidate label suppression term to ED-PLL (ablation) | ED-PLL-w/o-NC · MSRCv2 | accuracy: 54.28 -> 56.57 | Improved |
| add conflict-aware regularization to ED-PLL (ablation) | ED-PLL-w/o-NC · MSRCv2 | accuracy: 54.28 -> 57.04 | Improved |
| add label-weight update and candidate-evidence loss on top of original EDL loss (ablation) | ED-PLL-w/o-A · MSRCv2 | accuracy: 49.71 -> 54.28 | Improved |
| add both non-candidate and conflict terms (full ED-PLL) (ablation) | ED-PLL-w/o-NC · MNIST (q=0.3) | accuracy: 98.08 -> 99.16 | Improved |
| ED-PLL loss versus PiCO (structure-based) on harder benchmarks | ED-PLL · CIFAR-10 (q=0.5) | disambiguation accuracy: 91.35 -> 85.72 | Hurt |
| ED-PLL on Yahoo!News versus PiCO | ED-PLL · Yahoo!News | accuracy: 68.71 -> 62.34 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PiCO — PiCO: Contrastive Label Disambiguation for Partial Label Learning (ICLR 2022) · better on 5 of 7 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Partial Multi-Label Learning with Probabilistic Graphical Disambiguation (NeurIPS 2023 · similarity 0.90)
- FREDIS: A Fusion Framework of Refinement and Disambiguation for Unreliable Partial Label Learning (ICML 2023 · similarity 0.90)
- Adaptive Integration of Partial Label Learning and Negative Learning for Enhanced Noisy Label Learning (AAAI 2024 · similarity 0.89)
- Distilling Reliable Knowledge for Instance-Dependent Partial Label Learning (AAAI 2024 · similarity 0.89)
- Disentangled Partial Label Learning (AAAI 2024 · similarity 0.87)
- Conformal Prediction for Partial Label Learning (AAAI 2025 · 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.
