Research map / Label-efficient and robust learning / Recognition and long-tail
Meta-Cal: Well-controlled Post-hoc Calibration by Ranking
Xingchen Ma, Matthew B. Blaschko
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 |
|---|---|---|---|
| Augment temperature scaling with ranking-based miscoverage control | Meta-Cal on CIFAR10 · CIFAR10, DenseNet40 | ECE: 0.00510 -> 0.00434 | Improved |
| Augment temperature scaling with ranking-based coverage-accuracy control | Meta-Cal on CIFAR10 · CIFAR10, DenseNet40 | ECE: 0.00510 -> 0.00355 | Improved |
| Use coverage-accuracy control rather than miscoverage control | Meta-Cal · CIFAR10, DenseNet40 | ECE: 0.00434 -> 0.00355 | Improved |
| Use coverage-accuracy control rather than miscoverage control | Meta-Cal · CIFAR100, ResNet110 | ECE: 0.01815 -> 0.01441 | Improved |
| Use coverage-accuracy control rather than miscoverage control | Meta-Cal · CIFAR100, WideResNet32 | ECE: 0.01332 -> 0.01189 | Hurt |
| Apply Meta-Cal with either control variant instead of the strongest listed baseline | Meta-Cal · CIFAR100, ResNet110 | ECE: GPC 0.01348; MetaMis 0.01815; MetaAcc 0.01441 | Mixed |
| Apply Meta-Cal with coverage-accuracy control instead of temperature scaling | Meta-Cal · ImageNet, ResNet152 | ECE: 0.02023 -> 0.01138 | Improved |
Closest papers by meaning
- A Benchmark Study on Calibration (ICLR 2024 · similarity 0.87)
- Calibration Bottleneck: Over-compressed Representations are Less Calibratable (ICML 2024 · similarity 0.87)
- Confidence Calibration of Classifiers with Many Classes (NeurIPS 2024 · similarity 0.87)
- Improving Multi-Class Calibration through Normalization-Aware Isotonic Techniques (ICML 2025 · similarity 0.87)
- Divide et Calibra: Multiclass Local Calibration via Vector Quantization (NeurIPS 2026 · similarity 0.87)
- Reassessing How to Compare and Improve the Calibration of Machine Learning Models (ICLR 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.
