arXiv Machine Learning September 21, 2026 By Dane Williamson, Yangfeng Ji, Matthew Dwyer

CASE: Contrastive Activation for Class-Sensitive Explanations

Read the original on arXiv Machine Learning →

The paper introduces a diagnostic test for class sensitivity in saliency methods, revealing that many popular techniques produce nearly identical explanations regardless of the predicted class. This limitation appears across different architectures and datasets, indicating a structural issue. To address this, the authors propose CASE, a contrastive explanation method that isolates features uniquely discriminative for the predicted class, and demonstrate its improved fidelity and class specificity through experiments.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv AI

Jun 10

Visual-TCAV: Concept-based Attribution and Saliency Maps for Post-hoc Explainability in Image Classification

arXiv:2411. 05698v3 Announce Type: replace-cross Abstract: Convolutional Neural Networks (CNNs) have shown remarkable performance in image classification.

By Antonio De Santis, Riccardo Campi, Matteo Bianchi, Marco Brambilla

computer-vision benchmarks safety

arXiv Machine Learning

Jul 31

Contrastive Concept Importance: Explaining Pairwise Class Decisions Through Automatically Extracted Concept Representations

arXiv:2607. 27904v1 Announce Type: new Abstract: Concept-based explanations are a prevalent way to explain the decisions of complex black-box methods through semantically meaningful, human-interpretable concepts.

By Roel Visser, Isaac Roberts, Barbara Hammer

arXiv Machine Learning

Sep 23

eXplaining to Learn (eX2L): Regularization Using Contrastive Visual Explanation Pairs for Distribution Shifts

The paper introduces eXplaining to Learn (eX2L), an interpretable framework that regularizes a classifier by penalizing similarity between Grad‑CAM maps of the main label classifier and a confounder classifier. This approach decorrelates confounding features from latent representations during training. On the Spawrious Many‑to‑Many Hard Challenge benchmark, eX2L outperforms the current state‑of‑the‑art by 5.49% in average accuracy and 10.90% in worst‑group accuracy, while also demonstrating functional domain invariance through explicit label‑nuisance decoupling.

By Paulo Mario P. Medina, Jose Marie Antonio Mi\~noza, Sebastian C. Iba\~nez

benchmarks safety

arXiv Machine Learning

Jun 8

OPTIMUS-Prime: Minimal and Sufficient Concept Explanations for Deep Vision Models

arXiv:2606. 07180v1 Announce Type: cross Abstract: The growing demand for transparency in automated decision-making has propelled eXplainable Artificial Intelligence (XAI) to the forefront of machine learning research.

By Arthur Hoarau, Chenrui Zhu, Vu Linh Nguyen

computer-vision benchmarks

arXiv AI

Aug 13

Class Activation Mapping in Explainable Computer Vision: A Method-Centered Review of CNN, Transformer, and Foundation-Model-Era Visual Explanations

arXiv:2608. 12299v1 Announce Type: cross Abstract: Class activation mapping (CAM) is one of the most widely used visual explanation families in explainable artificial intelligence.

By AmirHossein Eshghi, Hamid Saadatfar, Seyyed Ali Hoseini, AmirMohsen Eshghi, Siavash Arjomand Bigdel

llms computer-vision benchmarks

arXiv AI

Jul 13

Explaining is Harder Than Predicting Alone: Evaluating Concept-based Explanations of MLLMs as ICL Visual Classifiers

arXiv:2605. 28215v2 Announce Type: replace Abstract: In-context learning (ICL) enables multimodal large language models (MLLMs) to classify images from a few labelled examples.

By Carmen Quiles-Ram\'irez, Leticia L. Rodr\'iguez, Nicol\'as Martorell, Natalia D\'iaz-Rodr\'iguez

llms multimodal benchmarks

One email a morning, machine-written

Related stories

Visual-TCAV: Concept-based Attribution and Saliency Maps for Post-hoc Explainability in Image Classification

Contrastive Concept Importance: Explaining Pairwise Class Decisions Through Automatically Extracted Concept Representations

eXplaining to Learn (eX2L): Regularization Using Contrastive Visual Explanation Pairs for Distribution Shifts

OPTIMUS-Prime: Minimal and Sufficient Concept Explanations for Deep Vision Models

Class Activation Mapping in Explainable Computer Vision: A Method-Centered Review of CNN, Transformer, and Foundation-Model-Era Visual Explanations

Explaining is Harder Than Predicting Alone: Evaluating Concept-based Explanations of MLLMs as ICL Visual Classifiers