paper-with-me

Papers

Challenges in Evaluating Explanation Methods for Static and Evolving Data

2026-08-06 · Jerzy Stefanowski arxiv

This paper addresses the limitations of Explainable Artificial Intelligence (XAI) with respect to insufficient evaluation. They are illustrated through the DetoxAI image recognition system for bias detection and concept unlearning. Then, an example of a human-grounded evaluation of methods for explaining image classification is presented. The paper further explores methods for adapting explanations to evolving data streams with concept drift. Experiences with adapting counterfactuals for this problem are discussed. Finally it is related to the challenges of tracking the co-evolution of data, models, and explanations.\footnote{This paper has been accepted for a publication in J.Nalepa (ed) Explainable AI in Space. Proceedings of EASi 2026 Workshop at IJCAI-ECAI 2026 Bremen, Springer CCIS vol 3107 (2016).}

📄 PDF Abstract BibTeX arXiv:2608.06351

Code (0)

등록된 구현이 없습니다.

Tasks

Image ClassificationBias Detection

Similar Papers 제목 키워드 기반

Explaining GNN over Evolving Graphs using Information Flow

2021-11-19 · Yazheng Liu, Xi Zhang, Sihong Xie

Graphs are ubiquitous in many applications, such as social networks, knowledge graphs, smart grids, etc.. Graph neural networks (GNN) are the current state-of-the-art for these applications, and yet remain obscure to hum…

Knowledge Graphs

Fair and Explainable Credit-Scoring under Concept Drift: Adaptive Explanation Frameworks for Evolving Populations

2025-11-05 · Shivogo John arxiv

Evolving borrower behaviors, shifting economic conditions, and changing regulatory landscapes continuously reshape the data distributions underlying modern credit-scoring systems. Conventional explainability techniques, …

Variable Detection

EXACT: Towards a platform for empirically benchmarking Machine Learning model explanation methods

2024-05-20 · Benedict Clark, Rick Wilming, Artur Dox, Paul Eschenbach 외

The evolving landscape of explainable artificial intelligence (XAI) aims to improve the interpretability of intricate machine learning (ML) models, yet faces challenges in formalisation and empirical validation, being an…

BenchmarkingExplainable artificial intelligenceExplainable Artificial Intelligence (XAI)

Carpe Diem: On the Evaluation of World Knowledge in Lifelong Language Models

2023-11-14 · Yujin Kim, Jaehong Yoon, Seonghyeon Ye, Sangmin Bae 외

The dynamic nature of knowledge in an ever-changing world presents challenges for language models trained on static data; the model in the real world often requires not only acquiring new knowledge but also overwriting o…

Continual LearningQuestion AnsweringWorld Knowledge

Counterfactual Explanations Under Concept Drift

2026-05-17 · Marcin Kostrzewa, Jerzy Stefanowski, Maciej Zięba arxiv

Counterfactual explanations (CFEs) provide actionable recourse, but most methods assume a static framework with fixed data and a trained classifier. This assumption breaks in evolving data environments, such as data stre…