Finding a Balanced Degree of Automation for Summary Evaluation

TL;DR

This paper introduces a flexible semi-automatic to fully automatic evaluation framework for summarization that balances human judgment reliability with automation efficiency, improving correlation with human assessments.

Contribution

It proposes the Lite2Pyramid, Lite3Pyramid, and Lite2.xPyramid metrics, combining human-labeled units with automated semantic analysis for more reliable and cost-effective summary evaluation.

Findings

Lite2Pyramid achieves the best summary-level correlations.

Lite3Pyramid performs comparably to existing automatic metrics.

Lite2.xPyramid offers a trade-off between correlation and manual effort.

Abstract

Human evaluation for summarization tasks is reliable but brings in issues of reproducibility and high costs. Automatic metrics are cheap and reproducible but sometimes poorly correlated with human judgment. In this work, we propose flexible semiautomatic to automatic summary evaluation metrics, following the Pyramid human evaluation method. Semi-automatic Lite2Pyramid retains the reusable human-labeled Summary Content Units (SCUs) for reference(s) but replaces the manual work of judging SCUs' presence in system summaries with a natural language inference (NLI) model. Fully automatic Lite3Pyramid further substitutes SCUs with automatically extracted Semantic Triplet Units (STUs) via a semantic role labeling (SRL) model. Finally, we propose in-between metrics, Lite2.xPyramid, where we use a simple regressor to predict how well the STUs can simulate SCUs and retain SCUs that are more difficult to simulate, which provides a smooth transition and balance between automation and manual evaluation. Comparing to 15 existing metrics, we evaluate human-metric correlations on 3 existing meta-evaluation datasets and our newly-collected PyrXSum (with 100/10 XSum examples/systems). It shows that Lite2Pyramid consistently has the best summary-level correlations; Lite3Pyramid works better than or comparable to other automatic metrics; Lite2.xPyramid trades off small correlation drops for larger manual effort reduction, which can reduce costs for future data collection. Our code and data are publicly available at: https://github.com/ZhangShiyue/Lite2-3Pyramid