Sentence-level AI-generated text detection on MOSAIC PD
95.6Macro-F1SenFlow
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SenFlowType=TB, Evaluation Protocol=Unified2026.06 | 95.6 | 98.9 | 91.2 | |
| SenDetEXType=TB, Evaluation Protocol=Unified2026.06 | 94.3 | 98.5 | 88.5 | |
| SeqXGPTType=TB, Evaluation Protocol=Unified2026.06 | 86.3 | 95 | 72.8 | |
| POGERType=TB, Evaluation Protocol=Unified2026.06 | 85.4 | 93.8 | 71.5 | |
| BinocularsType=TF, Evaluation Protocol=Unified, Threshold tuning strategy=single threshold tuned on the combined validation set2026.06 | 67.7 | 76.5 | 35.6 | |
| Fast-DetectGPTType=TF, Evaluation Protocol=Unified, Threshold tuning strategy=single threshold tuned on the combined validation set2026.06 | 58.6 | 57.4 | 18.2 |