# Image Classification on CIFAR-10 (test) (Epsilon-Specific Accuracy)

Source: [Image Classification on CIFAR-10 (test) (Epsilon-Specific Accuracy)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-epsilon-specific-accuracy>)

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

Data updated: 2026-02-26T05:26:53.315Z

Evaluation of Image Classification performance on the CIFAR-10 test set using accuracy metrics measured under specific epsilon constraints (ε=0.4 and ε=0.6), likely related to robustness or calibration testing.

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) | CIFAR-10 (test) |
| Metric | Acc (ε=0.4) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Acc (ε=0.4)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-epsilon-specific-accuracy.md?metric=Acc%20(%CE%B5%3D0.4)>): Higher is better; unit: PERCENT.

- [Acc (ε=0.6)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-epsilon-specific-accuracy.md?metric=Acc%20(%CE%B5%3D0.6)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Acc (ε=0.4) | Result date | Report paper |
| --- | --- | --- | --- |
| CORES2 | 93.76 | Not recorded | 68fa698f1e209131cea22a5b |
| NLNL | 92.43 | Not recorded | 68fa698f1e209131cea22a5b |
| SELF | 91.13 | Not recorded | 68fa698f1e209131cea22a5b |
| Lq | 87.13 | Not recorded | 68fa698f1e209131cea22a5b |
| Forward T | 83.27 | Not recorded | 68fa698f1e209131cea22a5b |
| LDMI | 83.04 | Not recorded | 68fa698f1e209131cea22a5b |
| Cross Entropy | 81.88 | Not recorded | 68fa698f1e209131cea22a5b |
| MAE | 61.63 | Not recorded | 68fa698f1e209131cea22a5b |

### CORES2

Acc (ε=0.4): 93.76. Result ID: 698db6708b517eb433234429.

[Reporting paper](<https://www.sota2.com/research/paper/68fa698f1e209131cea22a5b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa698f1e209131cea22a5b.md>)

[Implementation](<https://github.com/UCSC-REAL/cores>)

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

### NLNL

Acc (ε=0.4): 92.43. Result ID: 698db6708b517eb433234427.

[Reporting paper](<https://www.sota2.com/research/paper/68fa698f1e209131cea22a5b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66371e209131cea0edb3.md>)

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

### SELF

Acc (ε=0.4): 91.13. Result ID: 698db6708b517eb433234428.

[Reporting paper](<https://www.sota2.com/research/paper/68fa698f1e209131cea22a5b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66761e209131cea10452.md>)

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

### Lq

Acc (ε=0.4): 87.13. Result ID: 698db6708b517eb433234425.

[Reporting paper](<https://www.sota2.com/research/paper/68fa698f1e209131cea22a5b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa651f1e209131cea088eb.md>)

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

### Forward T

Acc (ε=0.4): 83.27. Result ID: 698db6708b517eb433234424.

[Reporting paper](<https://www.sota2.com/research/paper/68fa698f1e209131cea22a5b.md>)

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

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

### LDMI

Acc (ε=0.4): 83.04. Result ID: 698db6708b517eb433234426.

[Reporting paper](<https://www.sota2.com/research/paper/68fa698f1e209131cea22a5b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa669b1e209131cea1159a.md>)

[Implementation](<https://github.com/Newbeeer/L_DMI>)

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

### Cross Entropy

Acc (ε=0.4): 81.88. Result ID: 698db6708b517eb433234422.

[Reporting paper](<https://www.sota2.com/research/paper/68fa698f1e209131cea22a5b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa631e1e209131ce9fcf46.md>)

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

### MAE

Acc (ε=0.4): 61.63. Result ID: 698db6708b517eb433234423.

[Reporting paper](<https://www.sota2.com/research/paper/68fa698f1e209131cea22a5b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63e81e209131cea0158c.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.
