# Image Classification on CIFAR100 (test) (IDN Robustness Levels 20-50%)

Source: [Image Classification on CIFAR100 (test) (IDN Robustness Levels 20-50%)](<https://www.sota2.com/research/sota/image-classification-on-cifar100-test-idn-robustness-levels-20-50-percent>)

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

Data updated: 2026-02-26T05:46:50.641Z

Evaluation of Image Classification performance on the CIFAR-100 test set, measuring robustness using the IDN metric across noise levels of 20%, 30%, 40%, 45%, and 50%.

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) | CIFAR100 (test) |
| Metric | IDN (20%) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [IDN (20%)](<https://www.sota2.com/research/sota/image-classification-on-cifar100-test-idn-robustness-levels-20-50-percent.md?metric=IDN%20(20%25)>): Higher is better; unit: PERCENT.

- [IDN (30%)](<https://www.sota2.com/research/sota/image-classification-on-cifar100-test-idn-robustness-levels-20-50-percent.md?metric=IDN%20(30%25)>): Higher is better; unit: PERCENT.

- [IDN (40%)](<https://www.sota2.com/research/sota/image-classification-on-cifar100-test-idn-robustness-levels-20-50-percent.md?metric=IDN%20(40%25)>): Higher is better; unit: PERCENT.

- [IDN (45%)](<https://www.sota2.com/research/sota/image-classification-on-cifar100-test-idn-robustness-levels-20-50-percent.md?metric=IDN%20(45%25)>): Higher is better; unit: PERCENT.

- [IDN (50%)](<https://www.sota2.com/research/sota/image-classification-on-cifar100-test-idn-robustness-levels-20-50-percent.md?metric=IDN%20(50%25)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | IDN (20%) | Result date | Report paper |
| --- | --- | --- | --- |
| CausalNL | 41.47 | Not recorded | 68fa6cb81e209131cea35949 |
| MentorNet | 38.91 | Not recorded | 68fa6cb81e209131cea35949 |
| Co-teaching | 37.96 | Not recorded | 68fa6cb81e209131cea35949 |
| T-Revision | 37.24 | Not recorded | 68fa6cb81e209131cea35949 |
| Reweight | 36.73 | Not recorded | 68fa6cb81e209131cea35949 |
| Decoupling | 36.53 | Not recorded | 68fa6cb81e209131cea35949 |
| Forward | 36.38 | Not recorded | 68fa6cb81e209131cea35949 |
| Mixup | 32.92 | Not recorded | 68fa6cb81e209131cea35949 |
| CE | 30.42 | Not recorded | 68fa6cb81e209131cea35949 |

### CausalNL

IDN (20%): 41.47. Result ID: 698f36373a6c314849f70cf4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea35949.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea35949.md>)

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

### MentorNet

IDN (20%): 38.91. Result ID: 698f36373a6c314849f70cef.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea35949.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64bf1e209131cea063ec.md>)

[Implementation](<https://github.com/google/mentornet>)

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

### Co-teaching

IDN (20%): 37.96. Result ID: 698f36373a6c314849f70ced.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea35949.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64fe1e209131cea07aa4.md>)

[Implementation](<https://github.com/bhanML/Co-teaching>)

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

### T-Revision

IDN (20%): 37.24. Result ID: 698f36373a6c314849f70cf3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea35949.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66d21e209131cea129ce.md>)

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

### Reweight

IDN (20%): 36.73. Result ID: 698f36373a6c314849f70cf2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea35949.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62011e209131ce9f7210.md>)

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

### Decoupling

IDN (20%): 36.53. Result ID: 698f36373a6c314849f70cee.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea35949.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa642f1e209131cea0344e.md>)

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

### Forward

IDN (20%): 36.38. Result ID: 698f36373a6c314849f70cf1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea35949.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63321e209131ce9fd4c8.md>)

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

### Mixup

IDN (20%): 32.92. Result ID: 698f36373a6c314849f70cf0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea35949.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa644e1e209131cea03e44.md>)

[Implementation](<https://github.com/facebookresearch/mixup-cifar10>)

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

### CE

IDN (20%): 30.42. Result ID: 698f36373a6c314849f70cec.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cb81e209131cea35949.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.
