Addressing Factual Error in Abstractive Dialogue Summarization via Span Identification and Correction

Faiz Ghifari Haznitrama, Young-Jun Lee, Ho-Jin Choi

Abstract: Abstractive dialogue summarization presents unique challenges due to the dynamic nature of conversations, involving multiple speakers, role changes, language variations, and informalities. Despite recent advancements in this field, summaries generated by existing methods often suffer from factual errors. To address this issue, post-processing correction has emerged as a promising approach that offers practicality and can be combined with other techniques. However, existing correction models still exhibit limitations, including false corrections that transform clean summaries into incorrect ones. We propose a simple and straightforward framework for correction with the main idea to separate the identification and use its results as guidance for better correction. Initially, the framework determines whether a summary contains factual errors and proceeds to identify the wrong part. This identified segment then serves as guidance for the correction. Our evaluation results demonstrated the effectiveness of our identifier and corrector model in terms of detecting incorrect summaries and performing corrections while highlighting its flexibility. Furthermore, the factuality human evaluation further emphasizes the ability of our approach to achieve accurate correction while preventing false correction.