# AI-generated text detection on PubMed Generated by GPT-4 (test)

Source: [AI-generated text detection on PubMed Generated by GPT-4 (test)](<https://www.sota2.com/research/sota/ai-generated-text-detection-on-pubmed-generated-by-gpt-4-test>)

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

Data updated: 2026-02-26T05:25:55.912Z

Evaluation of AI-generated text detection models on the test split of the PubMed dataset, specifically focusing on text generated by GPT-4, measured using AUROC.

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) | PubMed Generated by GPT-4 (test) |
| Metric | AUROC |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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/ai-generated-text-detection-on-pubmed-generated-by-gpt-4-test.md?metric=AUROC>): Higher is better; unit: SCALAR.

## Recorded results

| Method | AUROC | Result date | Report paper |
| --- | --- | --- | --- |
| DPIC | 0.9438 | Not recorded | 68fa73991e209131cea5d805 |
| DPIC | 0.9394 | Not recorded | 68fa73991e209131cea5d805 |
| Fast-DetectGPT | 0.8498 | Not recorded | 68fa73991e209131cea5d805 |
| Likelihood | 0.81 | Not recorded | 68fa73991e209131cea5d805 |
| RADAR | 0.8029 | Not recorded | 68fa73991e209131cea5d805 |
| LogRank | 0.7997 | Not recorded | 68fa73991e209131cea5d805 |
| DNA-GPT | 0.7163 | Not recorded | 68fa73991e209131cea5d805 |
| LRR | 0.681 | Not recorded | 68fa73991e209131cea5d805 |
| RoBERTa-large | 0.6067 | Not recorded | 68fa73991e209131cea5d805 |
| DNA-GPT | 0.5745 | Not recorded | 68fa73991e209131cea5d805 |
| DNA-GPT | 0.5366 | Not recorded | 68fa73991e209131cea5d805 |
| RoBERTa-base | 0.5309 | Not recorded | 68fa73991e209131cea5d805 |
| NPR | 0.4976 | Not recorded | 68fa73991e209131cea5d805 |
| DetectGPT | 0.3675 | Not recorded | 68fa73991e209131cea5d805 |
| Entropy | 0.3296 | Not recorded | 68fa73991e209131cea5d805 |

### DPIC

AUROC: 0.9438. Result ID: 6993042f3a6c3148491312f0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

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

### DPIC

AUROC: 0.9394. Result ID: 6993042f3a6c3148491312f1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

Recorded evaluation configuration: {"Auxiliary Model":"Vicuna-7b"}. This record does not assert independent reproduction.

### Fast-DetectGPT

AUROC: 0.8498. Result ID: 6993042f3a6c3148491312ef.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

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

Recorded evaluation configuration: {"Strategy":"Zero-shot"}. This record does not assert independent reproduction.

### Likelihood

AUROC: 0.81. Result ID: 6993042f3a6c3148491312e6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

Recorded evaluation configuration: {"Strategy":"Zero-shot"}. This record does not assert independent reproduction.

### RADAR

AUROC: 0.8029. Result ID: 6993042f3a6c3148491312e5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f81e209131cea4e73e.md>)

Recorded evaluation configuration: {"Strategy":"Supervised"}. This record does not assert independent reproduction.

### LogRank

AUROC: 0.7997. Result ID: 6993042f3a6c3148491312e8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

Recorded evaluation configuration: {"Strategy":"Zero-shot"}. This record does not assert independent reproduction.

### DNA-GPT

AUROC: 0.7163. Result ID: 6993042f3a6c3148491312ea.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

Recorded evaluation configuration: {"Auxiliary Model":"Neo-2.7"}. This record does not assert independent reproduction.

### LRR

AUROC: 0.681. Result ID: 6993042f3a6c3148491312e9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea4720c.md>)

Recorded evaluation configuration: {"Strategy":"Zero-shot"}. This record does not assert independent reproduction.

### RoBERTa-large

AUROC: 0.6067. Result ID: 6993042f3a6c3148491312e4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

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

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

Recorded evaluation configuration: {"Backbone":"RoBERTa-large"}. This record does not assert independent reproduction.

### DNA-GPT

AUROC: 0.5745. Result ID: 6993042f3a6c3148491312eb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

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

### DNA-GPT

AUROC: 0.5366. Result ID: 6993042f3a6c3148491312ec.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

Recorded evaluation configuration: {"Auxiliary Model":"Vicuna-7b"}. This record does not assert independent reproduction.

### RoBERTa-base

AUROC: 0.5309. Result ID: 6993042f3a6c3148491312e3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

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

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

Recorded evaluation configuration: {"Backbone":"RoBERTa-base"}. This record does not assert independent reproduction.

### NPR

AUROC: 0.4976. Result ID: 6993042f3a6c3148491312ed.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

Recorded evaluation configuration: {"Strategy":"Zero-shot"}. This record does not assert independent reproduction.

### DetectGPT

AUROC: 0.3675. Result ID: 6993042f3a6c3148491312ee.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

Recorded evaluation configuration: {"Strategy":"Zero-shot"}. This record does not assert independent reproduction.

### Entropy

AUROC: 0.3296. Result ID: 6993042f3a6c3148491312e7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73991e209131cea5d805.md>)

Recorded evaluation configuration: {"Strategy":"Zero-shot"}. This record does not assert independent reproduction.

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