Fine-grained and Explainable Factuality Evaluation for Multimodal Summarization

TL;DR

This paper introduces two explainable, fine-grained evaluation frameworks for assessing the factuality of multimodal summarization, applicable with or without reference summaries, improving the reliability of model outputs.

Contribution

The paper presents novel reference-based and reference-free frameworks for factuality evaluation in multimodal summarization, enhancing applicability and interpretability.

Findings

The frameworks show high correlation with existing metrics.

The reference-free framework broadens evaluation scenarios.

Experimental results confirm the effectiveness of the proposed methods.

Abstract

Multimodal summarization aims to generate a concise summary based on the input text and image. However, the existing methods potentially suffer from unfactual output. To evaluate the factuality of multimodal summarization models, we propose two fine-grained and explainable evaluation frameworks (FALLACIOUS) for different application scenarios, i.e. reference-based factuality evaluation framework and reference-free factuality evaluation framework. Notably, the reference-free factuality evaluation framework doesn't need ground truth and hence it has a wider application scenario. To evaluate the effectiveness of the proposed frameworks, we compute the correlation between our frameworks and the other metrics. The experimental results show the effectiveness of our proposed method. We will release our code and dataset via github.