# Uncertainty Calibration on CIFAR-F Source: [Uncertainty Calibration on CIFAR-F]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:41:21.345Z Evaluation of model uncertainty calibration performance on the CIFAR-F dataset, typically measured using Expected Calibration Error (ECE). A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Uncertainty Calibration | | Dataset (including any stated split) | CIFAR-F | | Metric | ECE | | Unit | PERCENT | | Direction | Lower is better | | Results loaded | 27 of 27 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [ECE](): Lower is better; unit: PERCENT. ## Recorded results | Method | ECE | Result date | Report paper | | --- | --- | --- | --- | | Knowledge-Transferring-based Temperature Scaling | 5.97 | Not recorded | 68fa6f071e209131cea43629 | | Knowledge-Transferring-based Temperature Scaling | 6.64 | Not recorded | 68fa6f071e209131cea43629 | | ETS | 6.8 | Not recorded | 68fa6f071e209131cea43629 | | GPC | 7.1 | Not recorded | 68fa6f071e209131cea43629 | | IROVA | 8.13 | Not recorded | 68fa6f071e209131cea43629 | | TS | 8.15 | Not recorded | 68fa6f071e209131cea43629 | | TS-IR | 8.42 | Not recorded | 68fa6f071e209131cea43629 | | SBC | 8.54 | Not recorded | 68fa6f071e209131cea43629 | | IR | 8.97 | Not recorded | 68fa6f071e209131cea43629 | | TS | 11.3 | Not recorded | 68fa6f071e209131cea43629 | | SBC | 11.4 | Not recorded | 68fa6f071e209131cea43629 | | GPC | 11.76 | Not recorded | 68fa6f071e209131cea43629 | | Knowledge-Transferring-based Temperature Scaling | 12.15 | Not recorded | 68fa6f071e209131cea43629 | | ETS | 12.21 | Not recorded | 68fa6f071e209131cea43629 | | TS-IR | 12.67 | Not recorded | 68fa6f071e209131cea43629 | | IROVA | 13.37 | Not recorded | 68fa6f071e209131cea43629 | | GPC | 14.18 | Not recorded | 68fa6f071e209131cea43629 | | IR | 14.39 | Not recorded | 68fa6f071e209131cea43629 | | ETS | 15.24 | Not recorded | 68fa6f071e209131cea43629 | | TS | 15.27 | Not recorded | 68fa6f071e209131cea43629 | | SBC | 15.45 | Not recorded | 68fa6f071e209131cea43629 | | TS-IR | 15.52 | Not recorded | 68fa6f071e209131cea43629 | | IR | 15.75 | Not recorded | 68fa6f071e209131cea43629 | | IROVA | 16.23 | Not recorded | 68fa6f071e209131cea43629 | | Base | 19.73 | Not recorded | 68fa6f071e209131cea43629 | | Base | 25.37 | Not recorded | 68fa6f071e209131cea43629 | | Base | 29.22 | Not recorded | 68fa6f071e209131cea43629 | ### Knowledge-Transferring-based Temperature Scaling ECE: 5.97. Result ID: 69907d0a3a6c31484900eb01. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"IF":10}. This record does not assert independent reproduction. ### Knowledge-Transferring-based Temperature Scaling ECE: 6.64. Result ID: 69907d0a3a6c31484900eaf8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"IF":50}. This record does not assert independent reproduction. ### ETS ECE: 6.8. Result ID: 69907d0a3a6c31484900eafb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":10}. This record does not assert independent reproduction. ### GPC ECE: 7.1. Result ID: 69907d0a3a6c31484900eb00. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":10}. This record does not assert independent reproduction. ### IROVA ECE: 8.13. Result ID: 69907d0a3a6c31484900eafe. [Reporting paper]() Recorded evaluation configuration: {"IF":10}. This record does not assert independent reproduction. ### TS ECE: 8.15. Result ID: 69907d0a3a6c31484900eafa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":10}. This record does not assert independent reproduction. ### TS-IR ECE: 8.42. Result ID: 69907d0a3a6c31484900eafc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":10}. This record does not assert independent reproduction. ### SBC ECE: 8.54. Result ID: 69907d0a3a6c31484900eaff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":10}. This record does not assert independent reproduction. ### IR ECE: 8.97. Result ID: 69907d0a3a6c31484900eafd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":10}. This record does not assert independent reproduction. ### TS ECE: 11.3. Result ID: 69907d0a3a6c31484900eaf1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":50}. This record does not assert independent reproduction. ### SBC ECE: 11.4. Result ID: 69907d0a3a6c31484900eaf6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":50}. This record does not assert independent reproduction. ### GPC ECE: 11.76. Result ID: 69907d0a3a6c31484900eaf7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":50}. This record does not assert independent reproduction. ### Knowledge-Transferring-based Temperature Scaling ECE: 12.15. Result ID: 69907d0a3a6c31484900eaef. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"IF":100}. This record does not assert independent reproduction. ### ETS ECE: 12.21. Result ID: 69907d0a3a6c31484900eaf2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":50}. This record does not assert independent reproduction. ### TS-IR ECE: 12.67. Result ID: 69907d0a3a6c31484900eaf3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":50}. This record does not assert independent reproduction. ### IROVA ECE: 13.37. Result ID: 69907d0a3a6c31484900eaf5. [Reporting paper]() Recorded evaluation configuration: {"IF":50}. This record does not assert independent reproduction. ### GPC ECE: 14.18. Result ID: 69907d0a3a6c31484900eaee. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":100}. This record does not assert independent reproduction. ### IR ECE: 14.39. Result ID: 69907d0a3a6c31484900eaf4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":50}. This record does not assert independent reproduction. ### ETS ECE: 15.24. Result ID: 69907d0a3a6c31484900eae9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":100}. This record does not assert independent reproduction. ### TS ECE: 15.27. Result ID: 69907d0a3a6c31484900eae8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":100}. This record does not assert independent reproduction. ### SBC ECE: 15.45. Result ID: 69907d0a3a6c31484900eaed. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":100}. This record does not assert independent reproduction. ### TS-IR ECE: 15.52. Result ID: 69907d0a3a6c31484900eaea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":100}. This record does not assert independent reproduction. ### IR ECE: 15.75. Result ID: 69907d0a3a6c31484900eaeb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"IF":100}. This record does not assert independent reproduction. ### IROVA ECE: 16.23. Result ID: 69907d0a3a6c31484900eaec. [Reporting paper]() Recorded evaluation configuration: {"IF":100}. This record does not assert independent reproduction. ### Base ECE: 19.73. Result ID: 69907d0a3a6c31484900eaf9. [Reporting paper]() Recorded evaluation configuration: {"IF":10}. This record does not assert independent reproduction. ### Base ECE: 25.37. Result ID: 69907d0a3a6c31484900eaf0. [Reporting paper]() Recorded evaluation configuration: {"IF":50}. This record does not assert independent reproduction. ### Base ECE: 29.22. Result ID: 69907d0a3a6c31484900eae7. [Reporting paper]() Recorded evaluation configuration: {"IF":100}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.