# Machine-Generated Text Detection on GPT-3.5 (full)

Source: [Machine-Generated Text Detection on GPT-3.5 (full)](<https://www.sota2.com/research/sota/machine-generated-text-detection-on-gpt-3-5-full>)

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

Data updated: 2026-02-26T05:29:49.653Z

Evaluation of models for detecting text generated by GPT-3.5, using the full dataset split, measured by Accuracy and F1 score.

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) | GPT-3.5 (full) |
| Metric | Accuracy |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/machine-generated-text-detection-on-gpt-3-5-full.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/machine-generated-text-detection-on-gpt-3-5-full.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| PECOLA | 99.95 | Not recorded | 68fa73fd1e209131cea5f0f0 |
| CE+Margin | 99.9 | Not recorded | 68fa73fd1e209131cea5f0f0 |
| RoBERTa | 99.89 | Not recorded | 68fa73fd1e209131cea5f0f0 |
| CE+SCL | 99.74 | Not recorded | 68fa73fd1e209131cea5f0f0 |
| CoCo | 99.44 | Not recorded | 68fa73fd1e209131cea5f0f0 |
| IT:Clust | 99.09 | Not recorded | 68fa73fd1e209131cea5f0f0 |
| Fast-Detect. | 94.72 | Not recorded | 68fa73fd1e209131cea5f0f0 |
| GLTR | 93.5 | Not recorded | 68fa73fd1e209131cea5f0f0 |
| DetectGPT | 90.8 | Not recorded | 68fa73fd1e209131cea5f0f0 |

### PECOLA

Accuracy: 99.95. Result ID: 699313473a6c3148491384a7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73fd1e209131cea5f0f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73fd1e209131cea5f0f0.md>)

[Implementation](<https://github.com/lsc-1/Pecola>)

Recorded evaluation configuration: {"Shot":"7000"}. This record does not assert independent reproduction.

### CE+Margin

Accuracy: 99.9. Result ID: 699313473a6c3148491384a2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73fd1e209131cea5f0f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ba51e209131cea2f89c.md>)

Recorded evaluation configuration: {"Shot":"7000"}. This record does not assert independent reproduction.

### RoBERTa

Accuracy: 99.89. Result ID: 699313473a6c31484913849f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73fd1e209131cea5f0f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66131e209131cea0e18b.md>)

[Implementation](<https://github.com/pytorch/fairseq>)

Recorded evaluation configuration: {"Shot":"7000"}. This record does not assert independent reproduction.

### CE+SCL

Accuracy: 99.74. Result ID: 699313473a6c3148491384a1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73fd1e209131cea5f0f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b41e209131cea2393a.md>)

Recorded evaluation configuration: {"Shot":"7000"}. This record does not assert independent reproduction.

### CoCo

Accuracy: 99.44. Result ID: 699313473a6c3148491384a4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73fd1e209131cea5f0f0.md>)

Recorded evaluation configuration: {"Shot":"7000"}. This record does not assert independent reproduction.

### IT:Clust

Accuracy: 99.09. Result ID: 699313473a6c3148491384a3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73fd1e209131cea5f0f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c141e209131cea32053.md>)

Recorded evaluation configuration: {"Shot":"7000"}. This record does not assert independent reproduction.

### Fast-Detect.

Accuracy: 94.72. Result ID: 699313473a6c3148491384a6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73fd1e209131cea5f0f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76341e209131cea6bcf2.md>)

Recorded evaluation configuration: {"Shot":"7000"}. This record does not assert independent reproduction.

### GLTR

Accuracy: 93.5. Result ID: 699313473a6c3148491384a0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73fd1e209131cea5f0f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65e11e209131cea0d18b.md>)

Recorded evaluation configuration: {"Shot":"7000"}. This record does not assert independent reproduction.

### DetectGPT

Accuracy: 90.8. Result ID: 699313473a6c3148491384a5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73fd1e209131cea5f0f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa701a1e209131cea49064.md>)

Recorded evaluation configuration: {"Shot":"7000"}. This record does not assert independent reproduction.

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