Assessing the Impact of Sequence Length Learning on Classification Tasks for Transformer Encoder Models
DOI:
https://doi.org/10.32473/flairs.37.1.135283Palabras clave:
Transformer, Text Classification, Bias, Explainability, Data AugmentationResumen
Classification algorithms using Transformer architectures can be affected by the sequence length learning problem whenever observations from different classes have a different length distribution. This problem causes models to use sequence length as a predictive feature instead of relying on important textual information. Although most public datasets are not affected by this problem, privately owned corpora for fields such as medicine and insurance may carry this data bias. The exploitation of this sequence length feature poses challenges throughout the value chain as these machine learning models can be used in critical applications. In this paper, we empirically expose this problem and present approaches to minimize its impacts.
Descargas
Publicado
Cómo citar
Número
Sección
Licencia
Derechos de autor 2024 Jean-Thomas Baillargeon, Luc Lamontagne
Esta obra está bajo una licencia internacional Creative Commons Atribución-NoComercial 4.0.