On the Disagreement in Perturbation-based xAI -- Benchmarking Perturbation Choices for Flood Detection from SAR Images

• URL: http://arxiv.org/abs/2607.14743v1
• Date: Thu, 16 Jul 2026 09:12:38 GMT
• Title: On the Disagreement in Perturbation-based xAI -- Benchmarking Perturbation Choices for Flood Detection from SAR Images
• Abstract summary: Perturbation-based xAI methods are widely used to analyze the behavior and predictions of deep learning models.<n>We focus on two key parameters of the perturbation pipeline, including the size and shape of the perturbed regions.<n>We evaluate their consistency across perturbation strategies and their faithfulness to the model's reasoning.
• Score: 8.85335952493336
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Perturbation-based xAI methods are widely used to analyze the behavior and predictions of deep learning models. By altering input regions and measuring the resulting changes in class probabilities with respect to the original image, they assign relevance scores and generate heatmaps that reflect each region's contribution to the prediction. Despite their apparent simplicity, however, perturbation-based methods are sensitive to parameter choices. In this work, we focus on two key parameters of the perturbation pipeline, namely the patch geometry, including the size and shape of the perturbed regions, and the perturbation type, defined by the replacement scheme. Grounded in the use case of flood detection from Synthetic Aperture Radar imagery, we conduct a comprehensive investigation of how relevance estimation changes under different perturbation settings. Beyond visual inspection of the resulting relevance maps, we evaluate their consistency across perturbation strategies and their faithfulness to the model's reasoning. We demonstrate how different perturbation choices can steer the resulting relevance maps, yielding ambiguous and even contradictory explanations. Our findings emphasize the importance of methodological settings in perturbation-based xAI. They underscore the need to carefully inspect and evaluate perturbation choices and to treat them as an integral part when interpreting explanations, ensuring a robust understanding of both the explanations and model predictions.

Related papers

• What Makes a Representation Good for Single-Cell Perturbation Prediction? [37.84894703250901]
  PerturbedVAE is a general framework designed to resolve this signal imbalance.<n>We provide an identifiability analysis that characterizes the conditions under which perturbation effects can be reliably recovered.<n>PerturbedVAE achieves state-of-the-art performance on a widely used benchmark multiple evaluation settings.
  arXiv  Detail & Related papers  (2026-05-19T04:30:11Z)
• K-Models: a Flexible and Interpretable Method for Ordinal Clustering with Application to Antigen-Antibody Interaction Profiles [65.21921642886367]
  K-Models is a novel framework that integrates ordinal constraints and estimates key underlying elements of the random process generating the observed functional profiles.<n>The proposed method is evaluated through simulations and real-world applications.
  arXiv  Detail & Related papers  (2026-05-14T13:35:44Z)
• Investigating Anisotropy in Visual Grounding under Controlled Counterfactual Perturbations [1.423280626666929]
  Grounding models are rarely evaluated under semantically mismatched captions.<n>This work examines whether embedding anisotropy contributes to counterfactual failures.
  arXiv  Detail & Related papers  (2026-05-09T17:54:31Z)
• VIKING: Deep variational inference with stochastic projections [48.946143517489496]
  Variational mean field approximations tend to struggle with contemporary overparametrized deep neural networks.<n>We propose a simple variational family that considers two independent linear subspaces of the parameter space.<n>This allows us to build a fully-correlated approximate posterior reflecting the overparametrization.
  arXiv  Detail & Related papers  (2025-10-27T15:38:35Z)
• Introspection in Learned Semantic Scene Graph Localisation [7.222321327403328]
  This work investigates how semantics influence localisation performance and robustness in a self-supervised, contrastive semantic localisation framework.<n>We conduct a thorough post-hoc introspection analysis to probe whether the model filters environmental noise and prioritises distinctive landmarks over routine clutter.<n>Overall, the results indicate that the model learns noise-robust, semantically salient relations about place definition, thereby enabling explainable registration under challenging visual and structural variations.
  arXiv  Detail & Related papers  (2025-10-08T14:21:45Z)
• Unifying Perplexing Behaviors in Modified BP Attributions through Alignment Perspective [61.5509267439999]
  We present a unified theoretical framework for methods like GBP, RectGrad, LRP, and DTD.<n>We demonstrate that they achieve input alignment by combining the weights of activated neurons.<n>This alignment improves the visualization quality and reduces sensitivity to weight randomization.
  arXiv  Detail & Related papers  (2025-03-14T07:58:26Z)
• Geo-Localization Based on Dynamically Weighted Factor-Graph [74.75763142610717]
  Feature-based geo-localization relies on associating features extracted from aerial imagery with those detected by the vehicle's sensors. This requires that the type of landmarks must be observable from both sources. We present a dynamically weighted factor graph model for the vehicle's trajectory estimation.
  arXiv  Detail & Related papers  (2023-11-13T12:44:14Z)
• Prediction under Latent Subgroup Shifts with High-Dimensional Observations [30.433078066683848]
  We introduce a new approach to prediction in graphical models with latent-shift adaptation. Our novel form of RPM identifies causal latent structure in the source environment, and adapts properly to predict in the target.
  arXiv  Detail & Related papers  (2023-06-23T12:26:24Z)
• Semantic Change Detection with Asymmetric Siamese Networks [71.28665116793138]
  Given two aerial images, semantic change detection aims to locate the land-cover variations and identify their change types with pixel-wise boundaries. This problem is vital in many earth vision related tasks, such as precise urban planning and natural resource management. We present an asymmetric siamese network (ASN) to locate and identify semantic changes through feature pairs obtained from modules of widely different structures.
  arXiv  Detail & Related papers  (2020-10-12T13:26:30Z)
• Learning Disentangled Representations with Latent Variation Predictability [102.4163768995288]
  This paper defines the variation predictability of latent disentangled representations. Within an adversarial generation process, we encourage variation predictability by maximizing the mutual information between latent variations and corresponding image pairs. We develop an evaluation metric that does not rely on the ground-truth generative factors to measure the disentanglement of latent representations.
  arXiv  Detail & Related papers  (2020-07-25T08:54:26Z)
• DANCE: Enhancing saliency maps using decoys [35.46266461621123]
  We propose a framework that improves the robustness of saliency methods by following a two-step procedure. First, we introduce a perturbation mechanism that subtly varies the input sample without changing its intermediate representations. Second, we compute saliency maps for perturbed samples and propose a new method to aggregate saliency maps.
  arXiv  Detail & Related papers  (2020-02-03T01:21:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.