# Robustness Verification on CIFAR (test)

Source: [Robustness Verification on CIFAR (test)](<https://www.sota2.com/research/sota/robustness-verification-on-cifar-test>)

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

Data updated: 2026-02-26T06:08:14.047Z

Evaluation of Robustness Verification performance on the test split of the CIFAR dataset.

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 | Robustness Verification |
| Dataset (including any stated split) | CIFAR (test) |
| Metric | Verification Rate |
| 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

- [Verification Rate](<https://www.sota2.com/research/sota/robustness-verification-on-cifar-test.md?metric=Verification%20Rate>): Higher is better; unit: PERCENT.

- [Latency (s)](<https://www.sota2.com/research/sota/robustness-verification-on-cifar-test.md?metric=Latency%20(s)>): Lower is better; unit: TIME.

## Recorded results

| Method | Verification Rate | Result date | Report paper |
| --- | --- | --- | --- |
| BICCOS | 62 | Not recorded | 68fa5cc91e209131ce9d452a |
| BICCOS | 62 | Not recorded | 68fa5cc91e209131ce9d452a |
| BICCOS | 56 | Not recorded | 68fa5cc91e209131ce9d452a |
| BICCOS | 49 | Not recorded | 68fa5cc91e209131ce9d452a |
| BICCOS | 48.5 | Not recorded | 68fa5cc91e209131ce9d452a |
| BICCOS | 48.5 | Not recorded | 68fa5cc91e209131ce9d452a |
| GCP-CROWN (MIP cuts) | 48.5 | Not recorded | 68fa5cc91e209131ce9d452a |
| B-CROWN | 45.5 | Not recorded | 68fa5cc91e209131ce9d452a |
| MN-BaB | 42.5 | Not recorded | 68fa5cc91e209131ce9d452a |

### BICCOS

Verification Rate: 62. Result ID: 698bb7e17a967eddb770d794.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Implementation](<https://github.com/Lemutisme/BICCOS>)

Recorded evaluation configuration: {"Model":"CNN-B-Adv","epsilon":"2/255","timeout threshold":"200s"}. This record does not assert independent reproduction.

### BICCOS

Verification Rate: 62. Result ID: 698bb7e17a967eddb770d795.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Implementation](<https://github.com/Lemutisme/BICCOS>)

Recorded evaluation configuration: {"Model":"CNN-B-Adv-4","epsilon":"2/255","timeout threshold":"200s"}. This record does not assert independent reproduction.

### BICCOS

Verification Rate: 56. Result ID: 698bb7e17a967eddb770d793.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Implementation](<https://github.com/Lemutisme/BICCOS>)

Recorded evaluation configuration: {"Model":"CNN-A-Mix-4","epsilon":"2/255","timeout threshold":"200s"}. This record does not assert independent reproduction.

### BICCOS

Verification Rate: 49. Result ID: 698bb7e17a967eddb770d790.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Implementation](<https://github.com/Lemutisme/BICCOS>)

Recorded evaluation configuration: {"Model":"CNN-A-Adv","epsilon":"2/255","timeout threshold":"200s"}. This record does not assert independent reproduction.

### BICCOS

Verification Rate: 48.5. Result ID: 698bb7e17a967eddb770d791.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Implementation](<https://github.com/Lemutisme/BICCOS>)

Recorded evaluation configuration: {"Model":"CNN-A-Adv-4","epsilon":"2/255","timeout threshold":"200s"}. This record does not assert independent reproduction.

### BICCOS

Verification Rate: 48.5. Result ID: 698bb7e17a967eddb770d792.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Implementation](<https://github.com/Lemutisme/BICCOS>)

Recorded evaluation configuration: {"Model":"CNN-A-Mix","epsilon":"2/255","timeout threshold":"200s"}. This record does not assert independent reproduction.

### GCP-CROWN (MIP cuts)

Verification Rate: 48.5. Result ID: 698bb7e17a967eddb770d797.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa59d41e209131ce9baab1.md>)

Recorded evaluation configuration: {"Model":"CNN-A-Adv","epsilon":"2/255","timeout threshold":"200s"}. This record does not assert independent reproduction.

### B-CROWN

Verification Rate: 45.5. Result ID: 698bb7e17a967eddb770d796.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

Recorded evaluation configuration: {"Model":"CNN-A-Adv","epsilon":"2/255","timeout threshold":"200s"}. This record does not assert independent reproduction.

### MN-BaB

Verification Rate: 42.5. Result ID: 698bb7e17a967eddb770d798.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5cc91e209131ce9d452a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c7a1e209131cea342fe.md>)

Recorded evaluation configuration: {"Model":"CNN-A-Adv","epsilon":"2/255","timeout threshold":"600s"}. This record does not assert independent reproduction.

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