Beyond the Veil of Similarity: Quantifying Semantic Continuity in Explainable AI
This work addresses the need for more reliable and transparent AI systems by providing a quantitative measure for evaluating XAI methods, though it appears incremental in nature.
The authors tackled the problem of measuring semantic continuity in Explainable AI (XAI) methods, proposing a novel metric to assess whether similar inputs yield similar explanations in image recognition tasks, with the goal of enhancing model interpretability and trustworthiness.
We introduce a novel metric for measuring semantic continuity in Explainable AI methods and machine learning models. We posit that for models to be truly interpretable and trustworthy, similar inputs should yield similar explanations, reflecting a consistent semantic understanding. By leveraging XAI techniques, we assess semantic continuity in the task of image recognition. We conduct experiments to observe how incremental changes in input affect the explanations provided by different XAI methods. Through this approach, we aim to evaluate the models' capability to generalize and abstract semantic concepts accurately and to evaluate different XAI methods in correctly capturing the model behaviour. This paper contributes to the broader discourse on AI interpretability by proposing a quantitative measure for semantic continuity for XAI methods, offering insights into the models' and explainers' internal reasoning processes, and promoting more reliable and transparent AI systems.