Attribution maps are one of the most established tools to explain the functioning of computer vision models. They assign importance scores to input features, indicating how relevant each feature is for the prediction of a deep neural network.

Unlike the seminal work of taskonomy that relies on a large number of annotations as supervision and is thus computationally cumbersome, the proposed approach requires no human annotations and imposes no constraints on the architectures of the networks.

The explainability of deep learning models remains a significant challenge, particularly in the medical domain where interpretable outputs are critical for clinical trust and transparency. Path attribution methods such as Integrated Gradients rely on a baseline representing the absence of relevant features ("missingness"). Commonly used baselines, such as all-zero inputs, are often semantically meaningless, especially in medical contexts. While alternative baseline choices have been explored, existing methods lack a principled approach to dynamically select baselines tailored to each input. In this work, we examine the notion of missingness in the medical context, analyze its implications for baseline selection, and introduce a counterfactual-guided approach to address the limitations of conventional baselines. We argue that a generated counterfactual (i.e. clinically "normal" variation of the pathological input) represents a more accurate representation of a meaningful absence of features. We use a Variational Autoencoder in our implementation, though our concept is model-agnostic and can be applied with any suitable counterfactual method. We evaluate our concept on three distinct medical data sets and empirically demonstrate that counterfactual baselines yield more faithful and medically relevant attributions, outperforming standard baseline choices as well as other related methods.

Changes in image statistics or distracting background elements are pitfalls that prevent generalization and real-world applicability of such control policies.

Building on a geometric understanding of these attacks presented in recent work, we identify Lipschitz continuity conditions on models' gradients that lead to robust gradient-based attributions, and observe that the smoothness of the model's decision surface is related to the transferability of attacks across multiple attribution methods.

This limits the interpretation and generates a deterministic output at test time. In this paper we aim to improve on the current feature attribution methods by developing a more interpretable model, and thus more meaningful feature attribution maps.

ImageNet class labels - the class labels are unusable. In the fine-tuning phase, the pruned network is fine-tuned for 10 epochs with batch size 180. We conduct experiments for structured pruning methods on ImageNet. We observe same tendencies in the results (Table 4). Our method outperforms naive compression in terms of maintaining the attribution maps.