# Image Classification on ImageNet noise level 95 16H (test)

Source: [Image Classification on ImageNet noise level 95 16H (test)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-noise-level-95-16h-test>)

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

Data updated: 2026-02-26T05:42:15.970Z

Evaluation of Image Classification performance on the ImageNet-16H dataset corrupted with Noise Level 95, specifically using the test split.

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 | Image Classification |
| Dataset (including any stated split) | ImageNet noise level 95 16H (test) |
| Metric | Error Rate |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 7 of 7 |
| 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

- [Error Rate](<https://www.sota2.com/research/sota/image-classification-on-imagenet-noise-level-95-16h-test.md?metric=Error%20Rate>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Error Rate | Result date | Report paper |
| --- | --- | --- | --- |
| P+L (MAP TS) | 7.89 | Not recorded | 68fa6af21e209131cea2b4a5 |
| P+L (Uncalibrated) | 8.02 | Not recorded | 68fa6af21e209131cea2b4a5 |
| P+L (IMax Binning) | 8.62 | Not recorded | 68fa6af21e209131cea2b4a5 |
| L+L | 9.01 | Not recorded | 68fa6af21e209131cea2b4a5 |
| P+L (Ensemble TS) | 9.32 | Not recorded | 68fa6af21e209131cea2b4a5 |
| VGG-19 | 12.58 | Not recorded | 68fa6af21e209131cea2b4a5 |
| Human | 14.07 | Not recorded | 68fa6af21e209131cea2b4a5 |

### P+L (MAP TS)

Error Rate: 7.89. Result ID: 698e45a5ffc3901de09c3692.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6af21e209131cea2b4a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6af21e209131cea2b4a5.md>)

[Implementation](<https://github.com/GavinKerrigan/conf_matrix_and_calibration>)

Recorded evaluation configuration: {"Backbone":"VGG-19","Number of seeds":25,"Labeled data points":"5067-5152","Calibration":"MAP TS"}. This record does not assert independent reproduction.

### P+L (Uncalibrated)

Error Rate: 8.02. Result ID: 698e45a5ffc3901de09c3691.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6af21e209131cea2b4a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6af21e209131cea2b4a5.md>)

[Implementation](<https://github.com/GavinKerrigan/conf_matrix_and_calibration>)

Recorded evaluation configuration: {"Backbone":"VGG-19","Number of seeds":25,"Labeled data points":"5067-5152"}. This record does not assert independent reproduction.

### P+L (IMax Binning)

Error Rate: 8.62. Result ID: 698e45a5ffc3901de09c3694.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6af21e209131cea2b4a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6af21e209131cea2b4a5.md>)

[Implementation](<https://github.com/GavinKerrigan/conf_matrix_and_calibration>)

Recorded evaluation configuration: {"Backbone":"VGG-19","Number of seeds":25,"Labeled data points":"5067-5152","Calibration":"IMax Binning"}. This record does not assert independent reproduction.

### L+L

Error Rate: 9.01. Result ID: 698e45a5ffc3901de09c3690.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6af21e209131cea2b4a5.md>)

Recorded evaluation configuration: {"Backbone":"VGG-19","Number of seeds":25,"Labeled data points":"5067-5152"}. This record does not assert independent reproduction.

### P+L (Ensemble TS)

Error Rate: 9.32. Result ID: 698e45a5ffc3901de09c3693.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6af21e209131cea2b4a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6af21e209131cea2b4a5.md>)

[Implementation](<https://github.com/GavinKerrigan/conf_matrix_and_calibration>)

Recorded evaluation configuration: {"Backbone":"VGG-19","Number of seeds":25,"Labeled data points":"5067-5152","Calibration":"Ensemble TS"}. This record does not assert independent reproduction.

### VGG-19

Error Rate: 12.58. Result ID: 698e45a5ffc3901de09c368f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6af21e209131cea2b4a5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61d01e209131ce9f62f5.md>)

[Implementation](<http://www.robots.ox.ac.uk/~vgg/research/very_deep/>)

Recorded evaluation configuration: {"Backbone":"VGG-19"}. This record does not assert independent reproduction.

### Human

Error Rate: 14.07. Result ID: 698e45a5ffc3901de09c368e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6af21e209131cea2b4a5.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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