Table 12 RoBERTa performance across text-length bins.
From: Classifying human vs. AI text with machine learning and explainable transformer models
Length Bin | N | Accuracy | Precision | Recall | F1-score |
|---|---|---|---|---|---|
Very Short | 3 | 1.000 | 1.000 | 1.000 | 1.000 |
Short | 11 | 1.000 | 1.000 | 1.000 | 1.000 |
Medium | 40 | 0.950 | 0.952 | 0.952 | 0.952 |
Long | 46 | 1.000 | 1.000 | 1.000 | 1.000 |
