# Membership Inference Attack on ImageNet-1k Source: [Membership Inference Attack on ImageNet-1k]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:58:49.566Z Evaluation of a model's susceptibility to Membership Inference Attacks on the ImageNet-1k dataset, typically measured by Precision at specific False Positive Rates (FPR). 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) | ImageNet-1k | | Metric | Precision @ 1% FPR | | 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 - [Precision @ 1% FPR](): Higher is better; unit: PERCENT. - [Precision @ 0.1% FPR](): Higher is better; unit: PERCENT. ## Recorded results | Method | Precision @ 1% FPR | Result date | Report paper | | --- | --- | --- | --- | | Quantile Regression Membership Inference Attack | 0.9745 | Not recorded | 68fa6ff91e209131cea48450 | | LiRA | 0.9457 | Not recorded | 68fa6ff91e209131cea48450 | | LiRA | 0.9374 | Not recorded | 68fa6ff91e209131cea48450 | | LiRA | 0.8911 | Not recorded | 68fa6ff91e209131cea48450 | | LiRA | 0.627 | Not recorded | 68fa6ff91e209131cea48450 | | Marginal | 0.4762 | Not recorded | 68fa6ff91e209131cea48450 | ### Quantile Regression Membership Inference Attack Precision @ 1% FPR: 0.9745. Result ID: 699103a03a6c3148490505af. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### LiRA Precision @ 1% FPR: 0.9457. Result ID: 699103a03a6c3148490505ae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50","n":8}. This record does not assert independent reproduction. ### LiRA Precision @ 1% FPR: 0.9374. Result ID: 699103a03a6c3148490505ad. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50","n":6}. This record does not assert independent reproduction. ### LiRA Precision @ 1% FPR: 0.8911. Result ID: 699103a03a6c3148490505ac. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50","n":4}. This record does not assert independent reproduction. ### LiRA Precision @ 1% FPR: 0.627. Result ID: 699103a03a6c3148490505ab. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50","n":2}. This record does not assert independent reproduction. ### Marginal Precision @ 1% FPR: 0.4762. Result ID: 699103a03a6c3148490505aa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.