# Property Inference (Senior) on Smile Detection Downstream 10000 samples (train)

Source: [Property Inference (Senior) on Smile Detection Downstream 10000 samples (train)](<https://www.sota2.com/research/sota/property-inference-senior-on-smile-detection-downstream-10000-samples-train>)

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

Data updated: 2026-02-26T05:25:19.709Z

Evaluation of Property Inference performance at the Senior level using the 10,000 sample training split of the Smile Detection Downstream dataset, measured via AUC ratios.

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 | Property Inference (Senior) |
| Dataset (including any stated split) | Smile Detection Downstream 10000 samples (train) |
| Metric | AUC (0.1% Ratio) |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 2 of 2 |
| 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

- [AUC (0.1% Ratio)](<https://www.sota2.com/research/sota/property-inference-senior-on-smile-detection-downstream-10000-samples-train.md?metric=AUC%20(0.1%25%20Ratio)>): Higher is better; unit: SCALAR.

- [AUC (1% Ratio)](<https://www.sota2.com/research/sota/property-inference-senior-on-smile-detection-downstream-10000-samples-train.md?metric=AUC%20(1%25%20Ratio)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | AUC (0.1% Ratio) | Result date | Report paper |
| --- | --- | --- | --- |
| Manipulated Upstream Model (Zero-Activation Attack) | 0.89 | Not recorded | 68fa6ed41e209131cea42764 |
| Normal Upstream Model | 0.59 | Not recorded | 68fa6ed41e209131cea42764 |

### Manipulated Upstream Model (Zero-Activation Attack)

AUC (0.1% Ratio): 0.89. Result ID: 699062483a6c314849002810.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ed41e209131cea42764.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ed41e209131cea42764.md>)

[Implementation](<https://github.com/yulongt23/Transfer-Inference>)

Recorded evaluation configuration: {"Upstream Task":"ImageNet Classification","Target Property":"Senior"}. This record does not assert independent reproduction.

### Normal Upstream Model

AUC (0.1% Ratio): 0.59. Result ID: 699062483a6c31484900280f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ed41e209131cea42764.md>)

Recorded evaluation configuration: {"Upstream Task":"ImageNet Classification","Target Property":"Senior"}. This record does not assert independent reproduction.

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