# Membership Inference Attack on VAR (belonging) vs ImageNet (non-belonging)

Source: [Membership Inference Attack on VAR (belonging) vs ImageNet (non-belonging)](<https://www.sota2.com/research/sota/membership-inference-attack-on-var-belonging-vs-imagenet-non-belonging>)

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

Data updated: 2026-06-30T07:23:05.504Z

Evaluation of a model's susceptibility to Membership Inference Attacks, specifically testing membership status against the VAR dataset compared to non-membership in the ImageNet dataset, using True Positive Rate at 1% False Positive Rate (TPR@1%FPR) as the primary metric.

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 | Membership Inference Attack |
| Dataset (including any stated split) | VAR (belonging) vs. ImageNet (non-belonging) |
| Metric | TPR@1%FPR |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [TPR@1%FPR](<https://www.sota2.com/research/sota/membership-inference-attack-on-var-belonging-vs-imagenet-non-belonging.md?metric=TPR%401%25FPR>): Higher is better; unit: PERCENT.

## Recorded results

| Method | TPR@1%FPR | Result date | Report paper |
| --- | --- | --- | --- |
| Ours | 100 | Not recorded | 6a434d70ba73165dbb7437fe |
| ICAS | 7.1 | Not recorded | 6a434d70ba73165dbb7437fe |
| CFG-Diff | 2.5 | Not recorded | 6a434d70ba73165dbb7437fe |

### Ours

TPR@1%FPR: 100. Result ID: 6a436ebefbc373592c93713a.

[Reporting paper](<https://www.sota2.com/research/paper/6a434d70ba73165dbb7437fe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a434d70ba73165dbb7437fe.md>)

Recorded evaluation configuration: {"Original Model":"VAR"}. This record does not assert independent reproduction.

### ICAS

TPR@1%FPR: 7.1. Result ID: 6a436ebefbc373592c937139.

[Reporting paper](<https://www.sota2.com/research/paper/6a434d70ba73165dbb7437fe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79eb1e209131cea7ff25.md>)

Recorded evaluation configuration: {"Original Model":"VAR"}. This record does not assert independent reproduction.

### CFG-Diff

TPR@1%FPR: 2.5. Result ID: 6a436ebefbc373592c937138.

[Reporting paper](<https://www.sota2.com/research/paper/6a434d70ba73165dbb7437fe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79a71e209131cea7eb1d.md>)

Recorded evaluation configuration: {"Original Model":"VAR"}. This record does not assert independent reproduction.

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