# Image Classification on CIFAR-10 (CA and ASR Metrics)

Source: [Image Classification on CIFAR-10 (CA and ASR Metrics)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-ca-and-asr-metrics>)

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

Data updated: 2026-06-19T08:33:42.380Z

Evaluation of standard Image Classification performance on CIFAR-10, reporting Classification Accuracy (CA) and Attack Success Rate (ASR).

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 |
| Metric | Classification Accuracy (%) |
| Unit | PERCENT |
| Direction | Higher 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

- [Classification Accuracy (%)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-ca-and-asr-metrics.md?metric=Classification%20Accuracy%20(%25)>): Higher is better; unit: PERCENT.

- [Attack Success Rate (%)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-ca-and-asr-metrics.md?metric=Attack%20Success%20Rate%20(%25)>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Classification Accuracy (%) | Result date | Report paper |
| --- | --- | --- | --- |
| Original Attack | 92.73 | Not recorded | 6a34cd18206542251d7c5f38 |
| Original Attack | 90.23 | Not recorded | 6a34cd18206542251d7c5f38 |
| Adaptive Attack | 61.23 | Not recorded | 6a34cd18206542251d7c5f38 |
| QVec | 59.02 | Not recorded | 6a34cd18206542251d7c5f38 |
| Adaptive Attack | 53.07 | Not recorded | 6a34cd18206542251d7c5f38 |
| QVec | 48.13 | Not recorded | 6a34cd18206542251d7c5f38 |

### Original Attack

Classification Accuracy (%): 92.73. Result ID: 6a34fec8fbc373592c3260d5.

[Reporting paper](<https://www.sota2.com/research/paper/6a34cd18206542251d7c5f38.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b311e209131cea2cb8f.md>)

[Implementation](<https://github.com/Secure-AI-Systems-Group/Qu-ANTI-zation>)

Recorded evaluation configuration: {"Quantization level":"INT4"}. This record does not assert independent reproduction.

### Original Attack

Classification Accuracy (%): 90.23. Result ID: 6a34fec8fbc373592c3260d2.

[Reporting paper](<https://www.sota2.com/research/paper/6a34cd18206542251d7c5f38.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b311e209131cea2cb8f.md>)

[Implementation](<https://github.com/Secure-AI-Systems-Group/Qu-ANTI-zation>)

Recorded evaluation configuration: {"Quantization level":"INT8"}. This record does not assert independent reproduction.

### Adaptive Attack

Classification Accuracy (%): 61.23. Result ID: 6a34fec8fbc373592c3260d3.

[Reporting paper](<https://www.sota2.com/research/paper/6a34cd18206542251d7c5f38.md>)

Recorded evaluation configuration: {"Quantization level":"INT8"}. This record does not assert independent reproduction.

### QVec

Classification Accuracy (%): 59.02. Result ID: 6a34fec8fbc373592c3260d4.

[Reporting paper](<https://www.sota2.com/research/paper/6a34cd18206542251d7c5f38.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a34cd18206542251d7c5f38.md>)

Recorded evaluation configuration: {"Quantization level":"INT8"}. This record does not assert independent reproduction.

### Adaptive Attack

Classification Accuracy (%): 53.07. Result ID: 6a34fec8fbc373592c3260d6.

[Reporting paper](<https://www.sota2.com/research/paper/6a34cd18206542251d7c5f38.md>)

Recorded evaluation configuration: {"Quantization level":"INT4"}. This record does not assert independent reproduction.

### QVec

Classification Accuracy (%): 48.13. Result ID: 6a34fec8fbc373592c3260d7.

[Reporting paper](<https://www.sota2.com/research/paper/6a34cd18206542251d7c5f38.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a34cd18206542251d7c5f38.md>)

Recorded evaluation configuration: {"Quantization level":"INT4"}. This record does not assert independent reproduction.

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