# AIGC Detection on BR-Gen (test) Source: [AIGC Detection on BR-Gen (test)]() Collection index: [llms.txt]() Data updated: 2026-03-11T07:54:21.213Z Evaluation of AIGC Detection performance on the test split of the BR-Gen dataset, reporting metrics including F1, AUC, and recall rates specific to GAN, Diffusion, Background, and Stuff generated content. 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 | AIGC Detection | | Dataset (including any stated split) | BR-Gen (test) | | Metric | Real Recall@50 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 7 of 7 | | 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 - [Real Recall@50](): Higher is better; unit: PERCENT. - [F1 Score](): Higher is better; unit: PERCENT. - [AUC](): Higher is better; unit: PERCENT. - [Recall@50](): Higher is better; unit: PERCENT. - [Recall@50 (GAN)](): Higher is better; unit: PERCENT. - [Recall@50 (Diffusion)](): Higher is better; unit: PERCENT. - [Recall@50 (Background)](): Higher is better; unit: PERCENT. - [Recall@50 (Stuff)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Real Recall@50 | Result date | Report paper | | --- | --- | --- | --- | | NFA-ViT | 0.992 | Not recorded | 68fa78241e209131cea76fb5 | | FatFormer | 0.99 | Not recorded | 68fa78241e209131cea76fb5 | | AIDE | 0.986 | Not recorded | 68fa78241e209131cea76fb5 | | NPR | 0.946 | Not recorded | 68fa78241e209131cea76fb5 | | DIRE | 0.939 | Not recorded | 68fa78241e209131cea76fb5 | | LGrad | 0.937 | Not recorded | 68fa78241e209131cea76fb5 | | FreqNet | 0.825 | Not recorded | 68fa78241e209131cea76fb5 | ### NFA-ViT Real Recall@50: 0.992. Result ID: 69b11faa6dbb7075b95b4c57. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"SegFormer (b2 image encoder, b0 noise encoder)"}. This record does not assert independent reproduction. ### FatFormer Real Recall@50: 0.99. Result ID: 69b11faa6dbb7075b95b4c55. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### AIDE Real Recall@50: 0.986. Result ID: 69b11faa6dbb7075b95b4c56. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### NPR Real Recall@50: 0.946. Result ID: 69b11faa6dbb7075b95b4c54. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DIRE Real Recall@50: 0.939. Result ID: 69b11faa6dbb7075b95b4c52. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LGrad Real Recall@50: 0.937. Result ID: 69b11faa6dbb7075b95b4c51. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### FreqNet Real Recall@50: 0.825. Result ID: 69b11faa6dbb7075b95b4c53. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.