Weakly- and Semi-supervised Evidence Extraction

TL;DR

This paper introduces methods for evidence extraction that effectively combine limited evidence annotations with abundant document-level labels, significantly improving performance in classification tasks.

Contribution

It presents novel semi-supervised techniques that leverage both weak and strong supervision for evidence extraction, outperforming existing interpretability baselines.

Findings

Methods outperform baselines on evidence extraction tasks.

Significant gains achieved with only hundreds of evidence annotations.

Approach is effective even with limited strong supervision.

Abstract

For many prediction tasks, stakeholders desire not only predictions but also supporting evidence that a human can use to verify its correctness. However, in practice, additional annotations marking supporting evidence may only be available for a minority of training examples (if available at all). In this paper, we propose new methods to combine few evidence annotations (strong semi-supervision) with abundant document-level labels (weak supervision) for the task of evidence extraction. Evaluating on two classification tasks that feature evidence annotations, we find that our methods outperform baselines adapted from the interpretability literature to our task. Our approach yields substantial gains with as few as hundred evidence annotations. Code and datasets to reproduce our work are available at https://github.com/danishpruthi/evidence-extraction.