Explainable Artificial Intelligence (XAI)

Derniers papiers

• All you need is SAMPAT

  Jayadeva, Madhur Aswani · 13 juillet 2026

  The current state of the art in AI/ML rests on deep neural architectures, which, in general, suffer from a lack of interpretability. Interpretability is crucial to gleaning insights while analyzing experimental data, where quantitative predictions may not be adequate for a scientist. We present a th…

• All Explanations are Wrong, But Many Are Useful: Exploring the Rashomon Explanation Set with Large Language Models

  Pan Li · 13 juillet 2026

  Explaining machine-learning models is increasingly important for decision-making and consumer trust, yet it is widely believed to come at a cost: existing Explainable AI (XAI) methods suffer from a persistent accuracy-explainability trade-off. We argue that this trade-off is not fundamental, but an …

• ConceptSMILE: Auditing the Trustworthiness of Concept-Based Explainable AI

  Mohadeseh Mollapour, Koorosh Aslansefat, Zeinab Dehghani, Bhupesh Kumar Mishra, Tejal Shah, Zhibao Mian · 13 juillet 2026

  Concept-based explainable artificial intelligence (AI) can make model reasoning more human-understandable, but concept-level outputs are not automatically trustworthy. We introduce ConceptSMILE, a model-agnostic perturbation-based auditing framework for evaluating the reliability of concept-based ex…

• How are linear representations learned? Exact solutions to the dynamics of abstraction

  William W. Yang, Andrew M. Saxe, Peter E. Latham · 13 juillet 2026

  In artificial and biological neural networks, concepts are often encoded as consistent linear directions in representation space. In deep learning, this idea is known as the linear representation hypothesis and underpins many interpretability and control methods based on linear probes, from concept …

• Steering Neural Network Training through Interpretable Constraints Based on Partial Dependence

  Yann Claes, Pierre Geurts, V\^an Anh Huynh-Thu · 10 juillet 2026

  Over the last few years, there has been an increased interest in making machine learning models more interpretable. Although a great deal of effort goes into developing techniques for interpreting the interactions learned by a given model, fewer studies focus on assessing the quality of such explana…

• CausalDS: Benchmarking Causal Reasoning in Data-Science Agents

  Andrej Leban, Yuekai Sun · 10 juillet 2026

  Large language models (LLMs) increasingly act as integrated data-science agents, combining abstract reasoning with advanced tool use. Yet the relevant benchmark landscape largely divides into symbolic causal reasoning benchmarks without realistic data analysis or data analysis benchmarks without a p…

• DeepPySR -- A Symbolic Regression Framework with Dynamic Pruning, Pareto Selection, and Hierarchical Composition for Real-World Scientific Discovery

  Fuling Chen, Kevin Vinsen, Phillip Melton, Rae-Chi Huang · 10 juillet 2026

  Symbolic regression (SR) discovers analytical equations from data, yielding glass-box models with directly interpretable formulas, unlike black-box methods that rely on unstable post-hoc tools such as SHAP or LIME. This transparency is crucial in clinical medicine and social science, but SR faces th…

• The Contribution of XAI for the Safe Development and Certification of AI: An Expert-Based Analysis

  Benjamin Fresz, Vincent Philipp G\"obels, Safa Omri, Danilo Brajovic, Andreas Aichele, Janika Kutz, Jens Neuh\"uttler, Marco F. Huber · 10 juillet 2026

  Developing and certifying safe - or so-called trustworthy - AI has become an increasingly salient issue, especially in light of upcoming regulation such as the EU AI Act. In this context, the black-box nature of machine learning models limits the use of conventional avenues of approach towards certi…

• Trustworthy Machine Learning through the Lens of Combinatorial Optimization: Survey and Research Perspectives

  Thibaut Vidal, Julien Ferry · 10 juillet 2026

  Modern machine learning (ML) increasingly relies on complex models whose behavior is difficult to characterize beyond empirical performance metrics. Across a wide range of tasks, including prediction, generation, and decision-making, models with similar empirical performance can exhibit markedly dif…

• Certified Interventional Fidelity: Anytime-Valid, Adaptive Evaluation of Causal Claims in Mechanistic Interpretability

  Amir Asiaee · 10 juillet 2026

  Mechanistic interpretability often evaluates explanations by intervening on a model: swapping hidden states, patching activations, ablating components, or comparing a compressed model to the original one. These experiments are usually summarized by a point estimate, even though the evaluation may be…

• Efficient Bayesian Deep Ensembles via Analytic Predictive Inference

  Sina Aghaee Dabaghan Fard, Marie Maros, Jaesung Lee · 9 juillet 2026

  We introduce an efficient Bayesian deep ensemble method for predictive regression designed to enhance interpretability while maintaining competitive predictive performance and computational efficiency. Our method combines the statistical rigor of Bayesian inference with the scalability of deep ensem…

• When Summaries Distort Decisions: Information Fidelity in LLM-Compressed Financial Analysis

  Hoyoung Lee, Suhwan Park, Seunghan Lee, Jun Seo, Jaehoon Lee, Sungdong Yoo, Minjae Kim, CheolWon Na, Zhangyang Wang, Zach Golkhou, Minkyu Kim, Sotirios Sabanis, Alejandro Lopez-Lira, Dhagash Mehta, Soonyoung Lee, Chanyeol Choi, Wonbin Ahn, Yongjae Lee · 9 juillet 2026

  Financial decision-makers face more information than they can directly inspect, making context compression necessary. Yet when large language models (LLMs) compress financial source material, they can alter the investment judgment supported by the original source. We frame this problem as informatio…

• Large Language Models (LLMs) and Generative AI in Cybersecurity and Privacy: A Survey of Dual-Use Risks, AI-Generated Malware, Explainability, and Defensive Strategies

  Kiarash Ahi, Saeed Valizadeh · 9 juillet 2026

  Large Language Models (LLMs) and generative AI (GenAI) systems, such as ChatGPT, Claude, Gemini, LLaMA, Copilot, Stable Diffusion by OpenAI, Anthropic, Google, Meta, Microsoft, Stability AI, respectively, are revolutionizing cybersecurity, enabling both automated defense and sophisticated attacks. T…

• ReMoDEx: A Local-to-Global Relevance-Based Model Decision Explainability Framework for large-Scale Image Datasets

  Abhay Kumar Pathak, Mrityunjay Chaubey, Manjari Gupta · 9 juillet 2026

  Deep learning image classifiers achieve strong predictive performance yet remain opaque in how decisions are formed. A model may predict correctly while relying on irrelevant cues, shortcut associations, peripheral structures, or device level artifacts instead of task relevant regions. On large scal…

• Riemannian Geometry for Pre-trained Language Model Embeddings

  Szczepan Konior, Alexandre Quemy, Przemys{\l}aw Klocek, Gr\'egoire Cattan, Bart{\l}omiej Sobieski · 9 juillet 2026

  Understanding the geometric structure of pre-trained language model embeddings matters for interpretability and safety. We ask whether sentence-level classification signal lives in the Riemannian geometry of contextual token embeddings, and probe it by extracting per-token pullback metrics from a le…

• Complexity-Budgeted, Interaction-Aware Interpretable Model for Tabular Data

  Srikumar Krishnamoorthy · 9 juillet 2026

  Inherently interpretable classifiers for tabular data typically rely on sparse features, rules, or patterns that users can inspect directly. The marginal feature-screening step common to these methods can discard variables whose predictive value emerges only through joint configurations with other v…

• Optimized Instance Alteration for Explaining and Assessing Robustness of Classifiers

  Evgenii Kuriabov, David Miller, Jia Li · 9 juillet 2026

  In this work, we propose a unified approach for diagnosing misclassification and assessing the robustness of black-box classifiers. Central to our method is an optimization framework that modifies an instance so that the classifier predicts a specified target label, while ensuring that the modificat…

• Mechanistic Interpretability for Neural Networks: Circuits, Sparse Features and Symbolic Reasoning

  Pranav Sawant, Jakub Krej\v{c}\'i · 9 juillet 2026

  This article offers a comprehensive overview of mechanistic interpretability, an emerging field that seeks to reverse-engineer the internal algorithms of modern neural networks. While traditional explainable AI methods often stop at surface-level input-output correlations, this approach directly add…

• Learning social norms enhances compatibility in dynamic human-AI coordination

  Yi Yang, Siyuan Liu, Xin Gao, Huamu Sun, Chao Liu, Qing Zhou, Bingbing Nie · 9 juillet 2026

  Humans continuously coordinate with others in dynamic interactions, often through implicit, hard-to-quantify social norms that act as shared tacit expectations among interacting agents. As AI agents, including large language models (LLMs), become embedded in daily life, they increasingly participate…

• ORCAID: Oblique Rule-Based Continuous-Action Interpretation for Deep RL Policies

  Ignacio D. Lopez-Miguel, Ezio Bartocci, Thomas Eiter, Martin Tappler · 9 juillet 2026

  Explainability remains a key issue in reinforcement learning (RL). Distilling an interpretable policy from an agent trained in a complex environment is particularly challenging when the action space is continuous. We introduce ORCAID, a novel method for extracting interpretable rule-based policies f…

• Driving the Wrong Way: Leveraging Interpretability in End2End Autonomous Driving Models

  Franz Motzkus, Sebastian Bernhard · 8 juillet 2026

  The increasing adoption of end-to-end learning for autonomous driving introduces increased model complexity and opacity, raising the risk of learning undesired or erroneous behavior. In this work, we integrate unsupervised dictionary learning as a post hoc interpretability module within state-of-the…

• From Graphs to Gradients: Physics-Inspired Structural Attribution for Cyber-Physical IoT Systems and Beyond

  Spyridon Evangelatos, Christos Diou, Georgios Th. Papadopoulos, Evangelos Markakis, Panagiotis Sarigiannidis · 8 juillet 2026

  Interpretable explanation methods in Artificial Intelligence aim to uncover the underlying causes and their effects, enabling a deeper understanding of why a system behaves in a certain way under different inputs. Unlike traditional explainability methods, which mainly highlight correlations between…

• More Convincing, Not More Correct: Self-Play Reward Hacking of Reference-Free LLM Judges

  Chenyu Zhou · 8 juillet 2026

  Training a language model against its own reference-free judgments (the premise of self-rewarding, self-play, and LLM-as-a-judge pipelines) assumes a model's verdict on a shown answer tracks correctness. We show it fails structurally: conditioned on a candidate, a judge scores plausibility, not corr…

• Agent Step Value: Probing the Observer Effect in Black-Box Traces

  Andrew Zhang, Chengzhan Li · 8 juillet 2026

  Final-answer scores hide which agent transitions helped or harmed a trace. We introduce Agent Step Value (ASV), a replay framework that scores before/after states with a stateless LLM evaluator over a fixed candidate set. ASV reports entropy movement and Bayesian surprise measure belief movement, wh…

• ExplAIner: A Declarative Query Language for Explaining Classification Models

  Marcelo Arenas, Pablo Barcel\'o, Diego Bustamante, Jose Caraball, Mar\'ia Alejandra Schild, Bernardo Subercaseaux · 8 juillet 2026

  The XAI community has studied a wide range of queries and scores for explaining predictions of ML models. From a data management perspective, this proliferation of explanation notions calls for declarative query languages in which such notions can be specified, combined, and analyzed uniformly. In t…