# Risk Detection on MobileRisk

Source: [Risk Detection on MobileRisk](<https://www.sota2.com/research/sota/risk-detection-on-mobilerisk>)

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

Data updated: 2026-04-13T09:10:06.056Z

Evaluation of model performance on the MobileRisk dataset for general Risk Detection, reporting standard classification metrics (Accuracy, Precision, Recall, F1 Score) and False Positive Rate (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 | Risk Detection |
| Dataset (including any stated split) | MobileRisk |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/risk-detection-on-mobilerisk.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Precision](<https://www.sota2.com/research/sota/risk-detection-on-mobilerisk.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/risk-detection-on-mobilerisk.md?metric=Recall>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/risk-detection-on-mobilerisk.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [False Positive Rate](<https://www.sota2.com/research/sota/risk-detection-on-mobilerisk.md?metric=False%20Positive%20Rate>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| AutoGLM | 86.92 | Not recorded | 69dc786b7d0172140ec789a2 |
| Gemini-3 | 78.69 | Not recorded | 69dc786b7d0172140ec789a2 |
| AutoGLM | 76.47 | Not recorded | 69dc786b7d0172140ec789a2 |
| AutoGLM | 74.86 | Not recorded | 69dc786b7d0172140ec789a2 |
| GPT-5 | 68.92 | Not recorded | 69dc786b7d0172140ec789a2 |
| CORA | 61.27 | Not recorded | 69dc786b7d0172140ec789a2 |
| AutoGLM-VLM | 58.82 | Not recorded | 69dc786b7d0172140ec789a2 |
| UI-TARS-1.5 | 50 | Not recorded | 69dc786b7d0172140ec789a2 |

### AutoGLM

Accuracy: 86.92. Result ID: 69dcb29e47f34ce924c3db17.

[Reporting paper](<https://www.sota2.com/research/paper/69dc786b7d0172140ec789a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75921e209131cea68b5e.md>)

Recorded evaluation configuration: {"Safety Layer":"VLM-as-critic"}. This record does not assert independent reproduction.

### Gemini-3

Accuracy: 78.69. Result ID: 69dcb29e47f34ce924c3db12.

[Reporting paper](<https://www.sota2.com/research/paper/69dc786b7d0172140ec789a2.md>)

Recorded evaluation configuration: {"Type":"Base"}. This record does not assert independent reproduction.

### AutoGLM

Accuracy: 76.47. Result ID: 69dcb29e47f34ce924c3db16.

[Reporting paper](<https://www.sota2.com/research/paper/69dc786b7d0172140ec789a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75921e209131cea68b5e.md>)

Recorded evaluation configuration: {"Safety Layer":"Heuristic gate"}. This record does not assert independent reproduction.

### AutoGLM

Accuracy: 74.86. Result ID: 69dcb29e47f34ce924c3db15.

[Reporting paper](<https://www.sota2.com/research/paper/69dc786b7d0172140ec789a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75921e209131cea68b5e.md>)

Recorded evaluation configuration: {"Safety Layer":"Prompt"}. This record does not assert independent reproduction.

### GPT-5

Accuracy: 68.92. Result ID: 69dcb29e47f34ce924c3db11.

[Reporting paper](<https://www.sota2.com/research/paper/69dc786b7d0172140ec789a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695f39f053fdd5c361743ecb.md>)

Recorded evaluation configuration: {"Type":"Base"}. This record does not assert independent reproduction.

### CORA

Accuracy: 61.27. Result ID: 69dcb29e47f34ce924c3db18.

[Reporting paper](<https://www.sota2.com/research/paper/69dc786b7d0172140ec789a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69dc786b7d0172140ec789a2.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### AutoGLM-VLM

Accuracy: 58.82. Result ID: 69dcb29e47f34ce924c3db14.

[Reporting paper](<https://www.sota2.com/research/paper/69dc786b7d0172140ec789a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75921e209131cea68b5e.md>)

Recorded evaluation configuration: {"Type":"Base"}. This record does not assert independent reproduction.

### UI-TARS-1.5

Accuracy: 50. Result ID: 69dcb29e47f34ce924c3db13.

[Reporting paper](<https://www.sota2.com/research/paper/69dc786b7d0172140ec789a2.md>)

Recorded evaluation configuration: {"Type":"Base"}. This record does not assert independent reproduction.

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