# AI-generated text detection on M4GT-Bench All (test)

Source: [AI-generated text detection on M4GT-Bench All (test)](<https://www.sota2.com/research/sota/ai-generated-text-detection-on-m4gt-bench-all-test>)

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

Data updated: 2026-02-26T05:58:43.409Z

Evaluation of AI-generated text detection models on the M4GT-Bench All test set, reporting Precision, Recall, F1, and Accuracy.

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 | AI-generated text detection |
| Dataset (including any stated split) | M4GT-Bench All (test) |
| Metric | Precision |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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](<https://www.sota2.com/research/sota/ai-generated-text-detection-on-m4gt-bench-all-test.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/ai-generated-text-detection-on-m4gt-bench-all-test.md?metric=Recall>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/ai-generated-text-detection-on-m4gt-bench-all-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Accuracy](<https://www.sota2.com/research/sota/ai-generated-text-detection-on-m4gt-bench-all-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Precision | Result date | Report paper |
| --- | --- | --- | --- |
| RoBERTa | 96.96 | Not recorded | 6995478c09c33900fcd6d5da |
| DependencyAI | 96.02 | Not recorded | 6995478c09c33900fcd6d5da |
| XLM-R | 90.73 | Not recorded | 6995478c09c33900fcd6d5da |
| DependencyAI | 89.15 | Not recorded | 6995478c09c33900fcd6d5da |
| XLM-R | 86.64 | Not recorded | 6995478c09c33900fcd6d5da |
| Stylistic-SVM | 78.95 | Not recorded | 6995478c09c33900fcd6d5da |
| NELA-SVM | 64.5 | Not recorded | 6995478c09c33900fcd6d5da |
| GLTR-SVM | 52.81 | Not recorded | 6995478c09c33900fcd6d5da |
| GLTR-LR | 42.36 | Not recorded | 6995478c09c33900fcd6d5da |

### RoBERTa

Precision: 96.96. Result ID: 699ad87f5698665ec467eb76.

[Reporting paper](<https://www.sota2.com/research/paper/6995478c09c33900fcd6d5da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73d51e209131cea5e645.md>)

Recorded evaluation configuration: {"Training Protocol":"Leave-one-domain-out"}. This record does not assert independent reproduction.

### DependencyAI

Precision: 96.02. Result ID: 699ad87f5698665ec467eba8.

[Reporting paper](<https://www.sota2.com/research/paper/6995478c09c33900fcd6d5da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6995478c09c33900fcd6d5da.md>)

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

### XLM-R

Precision: 90.73. Result ID: 699ad87f5698665ec467eb77.

[Reporting paper](<https://www.sota2.com/research/paper/6995478c09c33900fcd6d5da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73d51e209131cea5e645.md>)

Recorded evaluation configuration: {"Training Protocol":"Leave-one-domain-out"}. This record does not assert independent reproduction.

### DependencyAI

Precision: 89.15. Result ID: 699ad87f5698665ec467eb7c.

[Reporting paper](<https://www.sota2.com/research/paper/6995478c09c33900fcd6d5da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6995478c09c33900fcd6d5da.md>)

Recorded evaluation configuration: {"Training Protocol":"Leave-one-domain-out"}. This record does not assert independent reproduction.

### XLM-R

Precision: 86.64. Result ID: 699ad87f5698665ec467eba7.

[Reporting paper](<https://www.sota2.com/research/paper/6995478c09c33900fcd6d5da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73d51e209131cea5e645.md>)

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

### Stylistic-SVM

Precision: 78.95. Result ID: 699ad87f5698665ec467eb7a.

[Reporting paper](<https://www.sota2.com/research/paper/6995478c09c33900fcd6d5da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73d51e209131cea5e645.md>)

Recorded evaluation configuration: {"Training Protocol":"Leave-one-domain-out"}. This record does not assert independent reproduction.

### NELA-SVM

Precision: 64.5. Result ID: 699ad87f5698665ec467eb7b.

[Reporting paper](<https://www.sota2.com/research/paper/6995478c09c33900fcd6d5da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73d51e209131cea5e645.md>)

Recorded evaluation configuration: {"Training Protocol":"Leave-one-domain-out"}. This record does not assert independent reproduction.

### GLTR-SVM

Precision: 52.81. Result ID: 699ad87f5698665ec467eb79.

[Reporting paper](<https://www.sota2.com/research/paper/6995478c09c33900fcd6d5da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73d51e209131cea5e645.md>)

Recorded evaluation configuration: {"Training Protocol":"Leave-one-domain-out"}. This record does not assert independent reproduction.

### GLTR-LR

Precision: 42.36. Result ID: 699ad87f5698665ec467eb78.

[Reporting paper](<https://www.sota2.com/research/paper/6995478c09c33900fcd6d5da.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73d51e209131cea5e645.md>)

Recorded evaluation configuration: {"Training Protocol":"Leave-one-domain-out"}. This record does not assert independent reproduction.

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