# Machine-generated text detection on WikiHow M4 benchmark (test)

Source: [Machine-generated text detection on WikiHow M4 benchmark (test)](<https://www.sota2.com/research/sota/machine-generated-text-detection-on-wikihow-m4-benchmark-test>)

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

Data updated: 2026-05-06T07:25:09.613Z

Evaluation of models for detecting machine-generated text using the test split of the WikiHow M4 benchmark dataset, reporting AUROC 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 | Machine-generated text detection |
| Dataset (including any stated split) | WikiHow M4 benchmark (test) |
| Metric | AUROC |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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

- [AUROC](<https://www.sota2.com/research/sota/machine-generated-text-detection-on-wikihow-m4-benchmark-test.md?metric=AUROC>): Higher is better; unit: PERCENT.

- [Accuracy](<https://www.sota2.com/research/sota/machine-generated-text-detection-on-wikihow-m4-benchmark-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AUROC | Result date | Report paper |
| --- | --- | --- | --- |
| RoBERTa | 100 | Not recorded | 68fa76fe1e209131cea704c7 |
| RoBERTa | 100 | Not recorded | 68fa76fe1e209131cea704c7 |
| RoBERTa | 100 | Not recorded | 68fa76fe1e209131cea704c7 |
| ExaGPT | 99.5 | Not recorded | 68fa76fe1e209131cea704c7 |
| DNA-GPT | 99.4 | Not recorded | 68fa76fe1e209131cea704c7 |
| ExaGPT | 98.8 | Not recorded | 68fa76fe1e209131cea704c7 |
| LR-GLTR | 97.5 | Not recorded | 68fa76fe1e209131cea704c7 |
| LR-GLTR | 94.8 | Not recorded | 68fa76fe1e209131cea704c7 |
| ExaGPT | 94.3 | Not recorded | 68fa76fe1e209131cea704c7 |
| DNA-GPT | 87.7 | Not recorded | 68fa76fe1e209131cea704c7 |
| LR-GLTR | 72.4 | Not recorded | 68fa76fe1e209131cea704c7 |
| DNA-GPT | 44.6 | Not recorded | 68fa76fe1e209131cea704c7 |

### RoBERTa

AUROC: 100. Result ID: 69faec5e47f34ce924dd74ad.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e721e209131cea3fdde.md>)

Recorded evaluation configuration: {"Generator":"ChatGPT"}. This record does not assert independent reproduction.

### RoBERTa

AUROC: 100. Result ID: 69faec5e47f34ce924dd74b1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e721e209131cea3fdde.md>)

Recorded evaluation configuration: {"Generator":"GPT-4"}. This record does not assert independent reproduction.

### RoBERTa

AUROC: 100. Result ID: 69faec5e47f34ce924dd74b5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e721e209131cea3fdde.md>)

Recorded evaluation configuration: {"Generator":"Dolly-v2"}. This record does not assert independent reproduction.

### ExaGPT

AUROC: 99.5. Result ID: 69faec5e47f34ce924dd74b0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

Recorded evaluation configuration: {"Generator":"ChatGPT"}. This record does not assert independent reproduction.

### DNA-GPT

AUROC: 99.4. Result ID: 69faec5e47f34ce924dd74af.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70cb1e209131cea4d2e1.md>)

Recorded evaluation configuration: {"Generator":"ChatGPT"}. This record does not assert independent reproduction.

### ExaGPT

AUROC: 98.8. Result ID: 69faec5e47f34ce924dd74b4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

Recorded evaluation configuration: {"Generator":"GPT-4"}. This record does not assert independent reproduction.

### LR-GLTR

AUROC: 97.5. Result ID: 69faec5e47f34ce924dd74ae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72711e209131cea56537.md>)

Recorded evaluation configuration: {"Generator":"ChatGPT"}. This record does not assert independent reproduction.

### LR-GLTR

AUROC: 94.8. Result ID: 69faec5e47f34ce924dd74b2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72711e209131cea56537.md>)

Recorded evaluation configuration: {"Generator":"GPT-4"}. This record does not assert independent reproduction.

### ExaGPT

AUROC: 94.3. Result ID: 69faec5e47f34ce924dd74b8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

Recorded evaluation configuration: {"Generator":"Dolly-v2"}. This record does not assert independent reproduction.

### DNA-GPT

AUROC: 87.7. Result ID: 69faec5e47f34ce924dd74b7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70cb1e209131cea4d2e1.md>)

Recorded evaluation configuration: {"Generator":"Dolly-v2"}. This record does not assert independent reproduction.

### LR-GLTR

AUROC: 72.4. Result ID: 69faec5e47f34ce924dd74b6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72711e209131cea56537.md>)

Recorded evaluation configuration: {"Generator":"Dolly-v2"}. This record does not assert independent reproduction.

### DNA-GPT

AUROC: 44.6. Result ID: 69faec5e47f34ce924dd74b3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76fe1e209131cea704c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70cb1e209131cea4d2e1.md>)

Recorded evaluation configuration: {"Generator":"GPT-4"}. This record does not assert independent reproduction.

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