Research map / Label-efficient and robust learning / Noisy labels
Decoupling "when to update" from "how to update"
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use Update by Disagreement instead of vanilla training | Face gender network · LFW Dataset #1, last iteration | balanced accuracy (mean): 88.9 ± 1.3 -> 92.2 ± 0.6 | Helped |
| Use Update by Disagreement instead of vanilla training | Face gender network · LFW Dataset #2, last iteration | balanced accuracy (mean): 92.8 ± 0.8 -> 93.7 ± 0.2 | Helped |
| Add Update by Disagreement on top of the s-model | s-model · LFW Dataset #1, last iteration | balanced accuracy (mean): 91.5 ± 0.4 -> 92.6 ± 0.1 | Helped |
| Add Update by Disagreement on top of the s-model | s-model · LFW Dataset #2, best iteration | balanced accuracy (mean): 94.1 ± 0.3 -> 94.8 ± 0.2 | Helped |
| Add Update by Disagreement on top of the s-model | s-model · LFW Dataset #2, last iteration | balanced accuracy (mean): 93.4 ± 0.4 -> 94.1 ± 0.4 | Helped |
Closest papers by meaning
- Learning to Learn From Noisy Labeled Data (CVPR 2019 · similarity 0.83)
- Mitigating Label Noise through Data Ambiguation (AAAI 2024 · similarity 0.80)
- Handling Label Noise via Instance-Level Difficulty Modeling and Dynamic Optimization (NeurIPS 2025 · similarity 0.80)
- Combating Semantic Contamination in Learning with Label Noise (AAAI 2025 · similarity 0.78)
- NoiseGPT: Label Noise Detection and Rectification through Probability Curvature (NeurIPS 2024 · similarity 0.78)
- Early Stopping Against Label Noise Without Validation Data (ICLR 2024 · similarity 0.78)
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.