We used Data Maps to model and characterize the AuTexTification dataset. This provides insights about the behaviour of individual samples during training across epochs (training dynamics). We characterized the samples across 3 dimensions: confidence, variability and correctness. This shows the presence of 3 regions: easy-to-learn, ambiguous and hard-to-learn examples. We used a classic CNN architecture and found out that training the model only on a subset of ambiguous examples improves the model's out-of-distribution generalization.
@article{arxiv.2405.11212,
title = {Automated Text Identification Using CNN and Training Dynamics},
author = {Claudiu Creanga and Liviu Petrisor Dinu},
journal= {arXiv preprint arXiv:2405.11212},
year = {2024}
}