Don't Waste a Single Annotation: Improving Single-Label Classifiers Through Soft Labels

TL;DR

This paper proposes a soft label training approach for single-label classifiers that leverages annotator disagreement and confidence to improve model performance and calibration.

Contribution

It introduces a novel method that uses ambiguous annotations and annotator confidence to generate soft labels, enhancing classifier training.

Findings

Improved classifier accuracy on test sets.

Enhanced calibration of predicted probabilities.

Effective use of annotator disagreement information.

Abstract

In this paper, we address the limitations of the common data annotation and training methods for objective single-label classification tasks. Typically, when annotating such tasks annotators are only asked to provide a single label for each sample and annotator disagreement is discarded when a final hard label is decided through majority voting. We challenge this traditional approach, acknowledging that determining the appropriate label can be difficult due to the ambiguity and lack of context in the data samples. Rather than discarding the information from such ambiguous annotations, our soft label method makes use of them for training. Our findings indicate that additional annotator information, such as confidence, secondary label and disagreement, can be used to effectively generate soft labels. Training classifiers with these soft labels then leads to improved performance and calibration on the hard label test set.