Attribution Upsampling should Redistribute, Not Interpolate

TL;DR

This paper introduces USU, a novel semantic-aware upsampling method for attribution maps in explainable AI, addressing issues of corruption caused by traditional interpolation techniques.

Contribution

It formalizes the problem of attribution upsampling, proves the limitations of interpolation, and proposes USU as a principled, mass-preserving redistribution approach.

Findings

USU preserves attribution mass and importance ordering.

USU outperforms traditional interpolation in faithfulness and coherence.

Experiments confirm USU's effectiveness across multiple datasets.

Abstract

Attribution methods in explainable AI rely on upsampling techniques that were designed for natural images, not saliency maps. Standard bilinear and bicubic interpolation systematically corrupts attribution signals through aliasing, ringing, and boundary bleeding, producing spurious high-importance regions that misrepresent model reasoning. We identify that the core issue is treating attribution upsampling as an interpolation problem that operates in isolation from the model's reasoning, rather than a mass redistribution problem where model-derived semantic boundaries must govern how importance flows. We present Universal Semantic-Aware Upsampling (USU), a principled method that reformulates upsampling through ratio-form mass redistribution operators, provably preserving attribution mass and relative importance ordering. Extending the axiomatic tradition of feature attribution to upsampling, we formalize four desiderata for faithful upsampling and prove that interpolation structurally violates three of them. These same three force any redistribution operator into a ratio form; the fourth selects the unique potential within this family, yielding USU. Controlled experiments on models with known attribution priors verify USU's formal guarantees; evaluation across ImageNet, CIFAR-10, and CUB-200 confirms consistent faithfulness improvements and qualitatively superior, semantically coherent explanations.