Structural Compactness as a Complementary Criterion for Explanation Quality
This work addresses the problem of evaluating explanation quality for researchers and practitioners in interpretable machine learning, though it is incremental as it complements existing metrics rather than introducing a new paradigm.
The paper tackled the challenge of quantitatively assessing explanation legibility in attribution methods by introducing Minimum Spanning Tree Compactness (MST-C), a graph-based metric that captures geometric properties like spread and cohesion, and showed it reliably distinguishes between explanation methods and exposes structural differences between models.
In the evaluation of attribution quality, the quantitative assessment of explanation legibility is particularly difficult, as it is influenced by varying shapes and internal organization of attributions not captured by simple statistics. To address this issue, we introduce Minimum Spanning Tree Compactness (MST-C), a graph-based structural metric that captures higher-order geometric properties of attributions, such as spread and cohesion. These components are combined into a single score that evaluates compactness, favoring attributions with salient points spread across a small area and spatially organized into few but cohesive clusters. We show that MST-C reliably distinguishes between explanation methods, exposes fundamental structural differences between models, and provides a robust, self-contained diagnostic for explanation compactness that complements existing notions of attribution complexity.