Explainable Artificial Intelligence (XAI)

Derniers papiers

• RF-Agent: Automated Reward Function Design via Language Agent Tree Search

  Ning Gao, Xiuhui Zhang, Xingyu Jiang, Mukang You, Mohan Zhang, Yue Deng · 2 mars 2026

  Designing efficient reward functions for low-level control tasks is a challenging problem. Recent research aims to reduce reliance on expert experience by using Large Language Models (LLMs) with task information to generate dense reward functions. These methods typically rely on training results as …

• Joint Distribution-Informed Shapley Values for Sparse Counterfactual Explanations

  Lei You, Yijun Bian, Lele Cao · 2 mars 2026

  Counterfactual explanations (CE) aim to reveal how small input changes flip a model's prediction, yet many methods modify more features than necessary, reducing clarity and actionability. We introduce \emph{COLA}, a model- and generator-agnostic post-hoc framework that refines any given CE by comput…

• ReasonX: Declarative Reasoning on Explanations

  Laura State, Salvatore Ruggieri, Franco Turini · 2 mars 2026

  Explaining opaque Machine Learning (ML) models has become an increasingly important challenge. However, current eXplanation in AI (XAI) methods suffer several shortcomings, including insufficient abstraction, limited user interactivity, and inadequate integration of symbolic knowledge. We propose Re…

• GreenServ: Energy-Efficient Context-Aware Dynamic Routing for Multi-Model LLM Inference

  Thomas Ziller, Shashikant Ilager, Alessandro Tundo, Ezio Bartocci, Leonardo Mariani, Ivona Brandic · 2 mars 2026

  Large language models (LLMs) demonstrate remarkable capabilities, but their broad deployment is limited by significant computational resource demands, particularly energy consumption during inference. Static, one-model-fits-all inference strategies are often inefficient, as they do not exploit the d…

• Uncertainty-aware Language Guidance for Concept Bottleneck Models

  Yangyi Li, Mengdi Huai · 2 mars 2026

  Concept Bottleneck Models (CBMs) provide inherent interpretability by first mapping input samples to high-level semantic concepts, followed by a combination of these concepts for the final classification. However, the annotation of human-understandable concepts requires extensive expert knowledge an…

• Beyond Explainable AI (XAI): An Overdue Paradigm Shift and Post-XAI Research Directions

  Saleh Afroogh, Seyd Ishtiaque Ahmed, Petra Ahrweiler, David Alvarez-Melis, Mansur Maturidi Arief, Emilia Barakova, Falco J. Bargagli-Stoffi, Erdem Biyik, Hanjie Chen, Xiang 'Anthony' Chen, Robert Clements, Keeley Crockett, Amit Dhurandhar, Fethiye Irmak Dogan, Mollie Dollinger, Motahhare Eslami, Aldo A Faisal, Arya Farahi, Melanie Fernandez Pradie, Saadia Gabrie, Diego Garcia-Olano, Marzyeh Ghassemi, Shaona Ghosh, Hatice Gunes, Ehsan Hajiramezanali, Stefan Haufe, Biwei Huang, Angel Hwang, Md Tauhidul Islam, Junfeng Jiao, Amir-Hossein Karimi, Saber Kazeminasab, Anastasia Kuzminykh, William La Cava, Brian Y. Lim, Xiaofeng Liu, Mohammad R. K. Mofrad, Alicia Parrish, Maria Perez-Ortiz, Shriti Raj, Swabha Swayamdipta, Salmon Talebi, Kush R. Varshney, Mihaela Vorvoreanu, Lily Weng, Alice Xiang, Yiming Xu, Ding Zhao, Jieyu Zhao · 2 mars 2026

  This study provides a cross-disciplinary examination of Explainable Artificial Intelligence (XAI) approaches-focusing on deep neural networks (DNNs) and large language models (LLMs)-and identifies empirical and conceptual limitations in current XAI. We discuss critical symptoms that stem from deeper…

• A multimodal slice discovery framework for systematic failure detection and explanation in medical image classification

  Yixuan Liu, Kanwal K. Bhatia, Ahmed E. Fetit · 2 mars 2026

  Despite advances in machine learning-based medical image classifiers, the safety and reliability of these systems remain major concerns in practical settings. Existing auditing approaches mainly rely on unimodal features or metadata-based subgroup analyses, which are limited in interpretability and …

• Global Interpretability via Automated Preprocessing: A Framework Inspired by Psychiatric Questionnaires

  Eric V. Strobl · 2 mars 2026

  Psychiatric questionnaires are highly context sensitive and often only weakly predict subsequent symptom severity, which makes the prognostic relationship difficult to learn. Although flexible nonlinear models can improve predictive accuracy, their limited interpretability can erode clinical trust. …

• Synthesising Counterfactual Explanations via Label-Conditional Gaussian Mixture Variational Autoencoders

  Junqi Jiang, Francesco Leofante, Antonio Rago, Francesca Toni · 2 mars 2026

  Counterfactual explanations (CEs) provide recourse recommendations for individuals affected by algorithmic decisions. A key challenge is generating CEs that are robust against various perturbation types (e.g. input and model perturbations) while simultaneously satisfying other desirable properties. …

• ArgLLM-App: An Interactive System for Argumentative Reasoning with Large Language Models

  Adam Dejl, Deniz Gorur, Francesca Toni · 2 mars 2026

  Argumentative LLMs (ArgLLMs) are an existing approach leveraging Large Language Models (LLMs) and computational argumentation for decision-making, with the aim of making the resulting decisions faithfully explainable to and contestable by humans. Here we propose a web-based system implementing ArgLL…

• Veritas: Generalizable Deepfake Detection via Pattern-Aware Reasoning

  Hao Tan, Jun Lan, Zichang Tan, Ajian Liu, Chuanbiao Song, Senyuan Shi, Huijia Zhu, Weiqiang Wang, Jun Wan, Zhen Lei · 2 mars 2026

  Deepfake detection remains a formidable challenge due to the complex and evolving nature of fake content in real-world scenarios. However, existing academic benchmarks suffer from severe discrepancies from industrial practice, typically featuring homogeneous training sources and low-quality testing …

• Hierarchical Concept-based Interpretable Models

  Oscar Hill, Mateo Espinosa Zarlenga, Mateja Jamnik · 2 mars 2026

  Modern deep neural networks remain challenging to interpret due to the opacity of their latent representations, impeding model understanding, debugging, and debiasing. Concept Embedding Models (CEMs) address this by mapping inputs to human-interpretable concept representations from which tasks can b…

• GLUScope: A Tool for Analyzing GLU Neurons in Transformer Language Models

  Sebastian Gerstner, Hinrich Sch\"utze · 2 mars 2026

  We present GLUScope, an open-source tool for analyzing neurons in Transformer-based language models, intended for interpretability researchers. We focus on more recent models than previous tools do; specifically we consider gated activation functions such as SwiGLU. This introduces a new challenge: …

• Certified Circuits: Stability Guarantees for Mechanistic Circuits

  Alaa Anani, Tobias Lorenz, Bernt Schiele, Mario Fritz, Jonas Fischer · 27 février 2026

  Understanding how neural networks arrive at their predictions is essential for debugging, auditing, and deployment. Mechanistic interpretability pursues this goal by identifying circuits - minimal subnetworks responsible for specific behaviors. However, existing circuit discovery methods are brittle…

• Correcting Human Labels for Rater Effects in AI Evaluation: An Item Response Theory Approach

  Jodi M. Casabianca, Maggie Beiting-Parrish · 27 février 2026

  Human evaluations play a central role in training and assessing AI models, yet these data are rarely treated as measurements subject to systematic error. This paper integrates psychometric rater models into the AI pipeline to improve the reliability and validity of conclusions drawn from human judgm…

• FIRE: A Comprehensive Benchmark for Financial Intelligence and Reasoning Evaluation

  Xiyuan Zhang, Huihang Wu, Jiayu Guo, Zhenlin Zhang, Yiwei Zhang, Liangyu Huo, Xiaoxiao Ma, Jiansong Wan, Xuewei Jiao, Yi Jing, Jian Xie · 27 février 2026

  We introduce FIRE, a comprehensive benchmark designed to evaluate both the theoretical financial knowledge of LLMs and their ability to handle practical business scenarios. For theoretical assessment, we curate a diverse set of examination questions drawn from widely recognized financial qualificati…

• LoBoost: Fast Model-Native Local Conformal Prediction for Gradient-Boosted Trees

  Vagner Santos, Victor Coscrato, Luben Cabezas, Rafael Izbicki, Thiago Ramos · 27 février 2026

  Gradient-boosted decision trees are among the strongest off-the-shelf predictors for tabular regression, but point predictions alone do not quantify uncertainty. Conformal prediction provides distribution-free marginal coverage, yet split conformal uses a single global residual quantile and can be p…

• RLHFless: Serverless Computing for Efficient RLHF

  Rui Wei, Hanfei Yu, Shubham Jain, Yogarajan Sivakumar, Devesh Tiwari, Jian Li, Seung-Jong Park, Hao Wang · 27 février 2026

  Reinforcement Learning from Human Feedback (RLHF) has been widely applied to Large Language Model (LLM) post-training to align model outputs with human preferences. Recent models, such as DeepSeek-R1, have also shown RLHF's potential to improve LLM reasoning on complex tasks. In RL, inference and tr…

• Fair feature attribution for multi-output prediction: a Shapley-based perspective

  Umberto Biccari, Alain Ib\'a\~nez de Opakua, Jos\'e Mar\'ia Mato, \'Oscar Millet, Roberto Morales, Enrique Zuazua · 27 février 2026

  In this article, we provide an axiomatic characterization of feature attribution for multi-output predictors within the Shapley framework. While SHAP explanations are routinely computed independently for each output coordinate, the theoretical necessity of this practice has remained unclear. By exte…

• Slice and Explain: Logic-Based Explanations for Neural Networks through Domain Slicing

  Luiz Fernando Paulino Queiroz, Carlos Henrique Leit\~ao Cavalcante, Thiago Alves Rocha · 26 février 2026

  Neural networks (NNs) are pervasive across various domains but often lack interpretability. To address the growing need for explanations, logic-based approaches have been proposed to explain predictions made by NNs, offering correctness guarantees. However, scalability remains a concern in these met…

• Precedence-Constrained Decision Trees and Coverings

  Micha{\l} Szyfelbein, Dariusz Dereniowski · 26 février 2026

  This work considers a number of optimization problems and reductive relations between them. The two main problems we are interested in are the \emph{Optimal Decision Tree} and \emph{Set Cover}. We study these two fundamental tasks under precedence constraints, that is, if a test (or set) $X$ is a pr…

• fEDM+: A Risk-Based Fuzzy Ethical Decision Making Framework with Principle-Level Explainability and Pluralistic Validation

  Abeer Dyoub, Francesca A. Lisi · 26 février 2026

  In a previous work, we introduced the fuzzy Ethical Decision-Making framework (fEDM), a risk-based ethical reasoning architecture grounded in fuzzy logic. The original model combined a fuzzy Ethical Risk Assessment module (fERA) with ethical decision rules, enabled formal structural verification thr…

• 2-Step Agent: A Framework for the Interaction of a Decision Maker with AI Decision Support

  Otto Nyberg, Fausto Carcassi, Giovanni Cin\`a · 26 février 2026

  Across a growing number of fields, human decision making is supported by predictions from AI models. However, we still lack a deep understanding of the effects of adoption of these technologies. In this paper, we introduce a general computational framework, the 2-Step Agent, which models the effects…

• MINAR: Mechanistic Interpretability for Neural Algorithmic Reasoning

  Jesse He, Helen Jenne, Max Vargas, Davis Brown, Gal Mishne, Yusu Wang, Henry Kvinge · 26 février 2026

  The recent field of neural algorithmic reasoning (NAR) studies the ability of graph neural networks (GNNs) to emulate classical algorithms like Bellman-Ford, a phenomenon known as algorithmic alignment. At the same time, recent advances in large language models (LLMs) have spawned the study of mecha…

• Under the Influence: Quantifying Persuasion and Vigilance in Large Language Models

  Sasha Robinson, Kerem Oktar, Katherine M. Collins, Ilia Sucholutsky, Kelsey R. Allen · 26 février 2026

  With increasing integration of Large Language Models (LLMs) into areas of high-stakes human decision-making, it is important to understand the risks they introduce as advisors. To be useful advisors, LLMs must sift through large amounts of content, written with both benevolent and malicious intent, …