Factual Consistency Evaluation for Text Summarization via Counterfactual Estimation

TL;DR

This paper introduces a new automatic evaluation metric for text summarization that effectively measures factual consistency by isolating the influence of language prior, showing improved correlation with human judgments without auxiliary tasks.

Contribution

It proposes a novel counterfactual estimation-based metric that isolates language prior effects, enhancing factual consistency evaluation in summarization without auxiliary tasks.

Findings

Improves correlation with human judgments

Effective across multiple datasets

Simplifies evaluation process

Abstract

Despite significant progress has been achieved in text summarization, factual inconsistency in generated summaries still severely limits its practical applications. Among the key factors to ensure factual consistency, a reliable automatic evaluation metric is the first and the most crucial one. However, existing metrics either neglect the intrinsic cause of the factual inconsistency or rely on auxiliary tasks, leading to an unsatisfied correlation with human judgments or increasing the inconvenience of usage in practice. In light of these challenges, we propose a novel metric to evaluate the factual consistency in text summarization via counterfactual estimation, which formulates the causal relationship among the source document, the generated summary, and the language prior. We remove the effect of language prior, which can cause factual inconsistency, from the total causal effect on the generated summary, and provides a simple yet effective way to evaluate consistency without relying on other auxiliary tasks. We conduct a series of experiments on three public abstractive text summarization datasets, and demonstrate the advantages of the proposed metric in both improving the correlation with human judgments and the convenience of usage. The source code is available at https://github.com/xieyxclack/factual_coco.