Contrastive Learning for Authorship Verification

Our results show that contrastive learning outperforms a classification-based approach to authorship verification under the tested settings. We identify loss function, batch size, training duration, pre-trained model, input context length, and random text span data augmentation as important factors of model performance. Based on these considerations, we develop a ModernBERT Bi-Encoder model that achieves 98.4% accuracy on the PAN21 authorship verification task.

Source: arXiv:2609.28471v1 - http://arxiv.org/abs/2609.28471v1 PDF: https://arxiv.org/pdf/2609.28471v1 Original Link: http://arxiv.org/abs/2609.28471v1