# Confidence Calibration on ImageNet ID (test)

Source: [Confidence Calibration on ImageNet ID (test)](<https://www.sota2.com/research/sota/confidence-calibration-on-imagenet-id-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:41:22.938Z

Evaluation of model confidence calibration performance on the test split of the ImageNet dataset using the In-Distribution (ID) configuration, 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 | Confidence Calibration |
| Dataset (including any stated split) | ImageNet ID (test) |
| Metric | ECE |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 6 of 6 |
| 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](<https://www.sota2.com/research/sota/confidence-calibration-on-imagenet-id-test.md?metric=ECE>): Lower is better; unit: PERCENT.

## Recorded results

| Method | ECE | Result date | Report paper |
| --- | --- | --- | --- |
| Base Classifier | 2.05 | Not recorded | 697c3afba8b01e76cc8da413 |
| ACET | 2.64 | Not recorded | 697c3afba8b01e76cc8da413 |
| OTIS | 2.71 | Not recorded | 697c3afba8b01e76cc8da413 |
| CEDA | 3.07 | Not recorded | 697c3afba8b01e76cc8da413 |
| VOS | 4.26 | Not recorded | 697c3afba8b01e76cc8da413 |
| CODEs | 4.75 | Not recorded | 697c3afba8b01e76cc8da413 |

### Base Classifier

ECE: 2.05. Result ID: 69977cec5698665ec4d16389.

[Reporting paper](<https://www.sota2.com/research/paper/697c3afba8b01e76cc8da413.md>)

Recorded evaluation configuration: {"Training Strategy":"None"}. This record does not assert independent reproduction.

### ACET

ECE: 2.64. Result ID: 69977cec5698665ec4d1638b.

[Reporting paper](<https://www.sota2.com/research/paper/697c3afba8b01e76cc8da413.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c01e209131cea0c2b9.md>)

Recorded evaluation configuration: {"Training Strategy":"OOD overconfidence mitigation"}. This record does not assert independent reproduction.

### OTIS

ECE: 2.71. Result ID: 69977cec5698665ec4d1638e.

[Reporting paper](<https://www.sota2.com/research/paper/697c3afba8b01e76cc8da413.md>)

[Method origin paper](<https://www.sota2.com/research/paper/697c3afba8b01e76cc8da413.md>)

Recorded evaluation configuration: {"Training Strategy":"OOD overconfidence mitigation"}. This record does not assert independent reproduction.

### CEDA

ECE: 3.07. Result ID: 69977cec5698665ec4d1638a.

[Reporting paper](<https://www.sota2.com/research/paper/697c3afba8b01e76cc8da413.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c01e209131cea0c2b9.md>)

Recorded evaluation configuration: {"Training Strategy":"OOD overconfidence mitigation"}. This record does not assert independent reproduction.

### VOS

ECE: 4.26. Result ID: 69977cec5698665ec4d1638d.

[Reporting paper](<https://www.sota2.com/research/paper/697c3afba8b01e76cc8da413.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c831e209131cea346d4.md>)

Recorded evaluation configuration: {"Training Strategy":"OOD overconfidence mitigation"}. This record does not assert independent reproduction.

### CODEs

ECE: 4.75. Result ID: 69977cec5698665ec4d1638c.

[Reporting paper](<https://www.sota2.com/research/paper/697c3afba8b01e76cc8da413.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a921e209131cea28faf.md>)

Recorded evaluation configuration: {"Training Strategy":"OOD overconfidence mitigation"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.