Sentence-level AI-generated text detection on MOSAIC PK
94.2Macro-F1SenFlow
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SenFlowType=TB, Evaluation Protocol=Unified2026.06 | 94.2 | 98.1 | 88.5 | |
| SenDetEXType=TB, Evaluation Protocol=Unified2026.06 | 92.5 | 97.7 | 85 | |
| SeqXGPTType=TB, Evaluation Protocol=Unified2026.06 | 76.2 | 88.3 | 52.9 | |
| POGERType=TB, Evaluation Protocol=Unified2026.06 | 71.9 | 84 | 44 | |
| BinocularsType=TF, Evaluation Protocol=Unified, Threshold tuning strategy=single threshold tuned on the combined validation set2026.06 | 71.7 | 79.3 | 43.4 | |
| Fast-DetectGPTType=TF, Evaluation Protocol=Unified, Threshold tuning strategy=single threshold tuned on the combined validation set2026.06 | 60.4 | 62.5 | 21.5 |