Explainable Artificial Intelligence (XAI)

Derniers papiers

• Aurora: Neuro-Symbolic AI Driven Advising Agent

  Lorena Amanda Quincoso Lugones, Christopher Kverne, Nityam Sharadkumar Bhimani, Ana Carolina Oliveira, Agoritsa Polyzou, Christine Lisetti, Janki Bhimani · 23 février 2026

  Academic advising in higher education is under severe strain, with advisor-to-student ratios commonly exceeding 300:1. These structural bottlenecks limit timely access to guidance, increase the risk of delayed graduation, and contribute to inequities in student support. We introduce Aurora, a modula…

• Learning Performance Maximizing Ensembles with Explainability Guarantees

  Vincent Pisztora, Jia Li · 23 février 2026

  In this paper we propose a method for the optimal allocation of observations between an intrinsically explainable glass box model and a black box model. An optimal allocation being defined as one which, for any given explainability level (i.e. the proportion of observations for which the explainable…

• Decomposing Representation Space into Interpretable Subspaces with Unsupervised Learning

  Xinting Huang, Michael Hahn · 23 février 2026

  Understanding internal representations of neural models is a core interest of mechanistic interpretability. Due to its large dimensionality, the representation space can encode various aspects about inputs. To what extent are different aspects organized and encoded in separate subspaces? Is it possi…

• Epistemic Traps: Rational Misalignment Driven by Model Misspecification

  Xingcheng Xu, Jingjing Qu, Qiaosheng Zhang, Chaochao Lu, Yanqing Yang, Na Zou, Xia Hu · 23 février 2026

  The rapid deployment of Large Language Models and AI agents across critical societal and technical domains is hindered by persistent behavioral pathologies including sycophancy, hallucination, and strategic deception that resist mitigation via reinforcement learning. Current safety paradigms treat t…

• Causal Explanations for Image Classifiers

  Hana Chockler, David A. Kelly, Daniel Kroening, Youcheng Sun · 23 février 2026

  Existing algorithms for explaining the output of image classifiers use different definitions of explanations and a variety of techniques to find them. However, none of the existing tools use a principled approach based on formal definitions of cause and explanation. In this paper we present a nove…

• Simplifying Outcomes of Language Model Component Analyses with ELIA

  Aaron Louis Eidt, Nils Feldhus · 23 février 2026

  While mechanistic interpretability has developed powerful tools to analyze the internal workings of Large Language Models (LLMs), their complexity has created an accessibility gap, limiting their use to specialists. We address this challenge by designing, building, and evaluating ELIA (Explainable L…

• Unifying Formal Explanations: A Complexity-Theoretic Perspective

  Shahaf Bassan, Xuanxiang Huang, Guy Katz · 23 février 2026

  Previous work has explored the computational complexity of deriving two fundamental types of explanations for ML model predictions: (1) *sufficient reasons*, which are subsets of input features that, when fixed, determine a prediction, and (2) *contrastive reasons*, which are subsets of input featur…

• Soft-CAM: Making black box models self-explainable for medical image analysis

  Kerol Djoumessi, Philipp Berens · 23 février 2026

  Convolutional neural networks (CNNs) are widely used for high-stakes applications like medicine, often surpassing human performance. However, most explanation methods rely on post-hoc attribution, approximating the decision-making process of already trained black-box models. These methods are often …

• Bayesian Optimality of In-Context Learning with Selective State Spaces

  Di Zhang, Jiaqi Xing · 23 février 2026

  We propose Bayesian optimal sequential prediction as a new principle for understanding in-context learning (ICL). Unlike interpretations framing Transformers as performing implicit gradient descent, we formalize ICL as meta-learning over latent sequence tasks. For tasks governed by Linear Gaussian S…

• Gradient Regularization Prevents Reward Hacking in Reinforcement Learning from Human Feedback and Verifiable Rewards

  Johannes Ackermann, Michael Noukhovitch, Takashi Ishida, Masashi Sugiyama · 23 février 2026

  Reinforcement Learning from Human Feedback (RLHF) or Verifiable Rewards (RLVR) are two key steps in the post-training of modern Language Models (LMs). A common problem is reward hacking, where the policy may exploit inaccuracies of the reward and learn an unintended behavior. Most previous works add…

• Provably Explaining Neural Additive Models

  Shahaf Bassan, Yizhak Yisrael Elboher, Tobias Ladner, Volkan \c{S}ahin, Jan Kretinsky, Matthias Althoff, Guy Katz · 20 février 2026

  Despite significant progress in post-hoc explanation methods for neural networks, many remain heuristic and lack provable guarantees. A key approach for obtaining explanations with provable guarantees is by identifying a cardinally-minimal subset of input features which by itself is provably suffici…

• Self-Evolving Multi-Agent Network for Industrial IoT Predictive Maintenance

  Rebin Saleh, Khanh Pham Dinh, Bal\'azs Vill\'anyi, Truong-Son Hy · 20 février 2026

  Industrial IoT predictive maintenance requires systems capable of real-time anomaly detection without sacrificing interpretability or demanding excessive computational resources. Traditional approaches rely on static, offline-trained models that cannot adapt to evolving operational conditions, while…

• Drones that Think on their Feet: Sudden Landing Decisions with Embodied AI

  Diego Ortiz Barbosa, Mohit Agrawal, Yash Malegaonkar, Luis Burbano, Axel Andersson, Gy\"orgy D\'an, Henrik Sandberg, Alvaro A. Cardenas · 20 février 2026

  Autonomous drones must often respond to sudden events, such as alarms, faults, or unexpected changes in their environment, that require immediate and adaptive decision-making. Traditional approaches rely on safety engineers hand-coding large sets of recovery rules, but this strategy cannot anticipat…

• RFEval: Benchmarking Reasoning Faithfulness under Counterfactual Reasoning Intervention in Large Reasoning Models

  Yunseok Han, Yejoon Lee, Jaeyoung Do · 20 février 2026

  Large Reasoning Models (LRMs) exhibit strong performance, yet often produce rationales that sound plausible but fail to reflect their true decision process, undermining reliability and trust. We introduce a formal framework for reasoning faithfulness, defined by two testable conditions: stance consi…

• Owen-based Semantics and Hierarchy-Aware Explanation (O-Shap)

  Xiangyu Zhou, Chenhan Xiao, Yang Weng · 20 février 2026

  Shapley value-based methods have become foundational in explainable artificial intelligence (XAI), offering theoretically grounded feature attributions through cooperative game theory. However, in practice, particularly in vision tasks, the assumption of feature independence breaks down, as features…

• Decoding the Human Factor: High Fidelity Behavioral Prediction for Strategic Foresight

  Ben Yellin, Ehud Ezra, Mark Foreman, Shula Grinapol · 20 février 2026

  Predicting human decision-making in high-stakes environments remains a central challenge for artificial intelligence. While large language models (LLMs) demonstrate strong general reasoning, they often struggle to generate consistent, individual-specific behavior, particularly when accurate predicti…

• When More Experts Hurt: Underfitting in Multi-Expert Learning to Defer

  Shuqi Liu, Yuzhou Cao, Lei Feng, Bo An, Luke Ong · 20 février 2026

  Learning to Defer (L2D) enables a classifier to abstain from predictions and defer to an expert, and has recently been extended to multi-expert settings. In this work, we show that multi-expert L2D is fundamentally more challenging than the single-expert case. With multiple experts, the classifier's…

• A feature-stable and explainable machine learning framework for trustworthy decision-making under incomplete clinical data

  Justyna Andrys-Olek, Paulina Tworek, Luca Gherardini, Mark W. Ruddock, Mary Jo Kurt, Peter Fitzgerald, Jose Sousa · 20 février 2026

  Machine learning models are increasingly applied to biomedical data, yet their adoption in high stakes domains remains limited by poor robustness, limited interpretability, and instability of learned features under realistic data perturbations, such as missingness. In particular, models that achieve…

• All Leaks Count, Some Count More: Interpretable Temporal Contamination Detection in LLM Backtesting

  Zeyu Zhang, Ryan Chen, Bradly C. Stadie · 20 février 2026

  To evaluate whether LLMs can accurately predict future events, we need the ability to \textit{backtest} them on events that have already resolved. This requires models to reason only with information available at a specified past date. Yet LLMs may inadvertently leak post-cutoff knowledge encoded du…

• Quantifying LLM Attention-Head Stability: Implications for Circuit Universality

  Karan Bali, Jack Stanley, Praneet Suresh, Danilo Bzdok · 20 février 2026

  In mechanistic interpretability, recent work scrutinizes transformer "circuits" - sparse, mono or multi layer sub computations, that may reflect human understandable functions. Yet, these network circuits are rarely acid-tested for their stability across different instances of the same deep learning…

• Beyond Linear Surrogates: High-Fidelity Local Explanations for Black-Box Models

  Sanjeev Shrestha, Rahul Dubey, Hui Liu · 20 février 2026

  With the increasing complexity of black-box machine learning models and their adoption in high-stakes areas, it is critical to provide explanations for their predictions. Existing local explanation methods lack in generating high-fidelity explanations. This paper proposes a novel local model agnosti…

• Better Think Thrice: Learning to Reason Causally with Double Counterfactual Consistency

  Victoria Lin, Xinnuo Xu, Rachel Lawrence, Risa Ueno, Amit Sharma, Javier Gonzalez, Niranjani Prasad · 20 février 2026

  Despite their strong performance on reasoning benchmarks, large language models (LLMs) have proven brittle when presented with counterfactual questions, suggesting weaknesses in their causal reasoning ability. While recent work has demonstrated that labeled counterfactual tasks can be useful benchma…

• CounterFlowNet: From Minimal Changes to Meaningful Counterfactual Explanations

  Oleksii Furman, Patryk Marsza{\l}ek, Jan Mas{\l}owski, Piotr Gai\'nski, Maciej Zi\k{e}ba, Marek \'Smieja · 20 février 2026

  Counterfactual explanations (CFs) provide human-interpretable insights into model's predictions by identifying minimal changes to input features that would alter the model's output. However, existing methods struggle to generate multiple high-quality explanations that (1) affect only a small portion…

• EduEVAL-DB: A Role-Based Dataset for Pedagogical Risk Evaluation in Educational Explanations

  Javier Irigoyen, Roberto Daza, Aythami Morales, Julian Fierrez, Francisco Jurado, Alvaro Ortigosa, Ruben Tolosana · 20 février 2026

  This work introduces EduEVAL-DB, a dataset based on teacher roles designed to support the evaluation and training of automatic pedagogical evaluators and AI tutors for instructional explanations. The dataset comprises 854 explanations corresponding to 139 questions from a curated subset of the Scien…

• Causality is Key for Interpretability Claims to Generalise

  Shruti Joshi, Aaron Mueller, David Klindt, Wieland Brendel, Patrik Reizinger, Dhanya Sridhar · 19 février 2026

  Interpretability research on large language models (LLMs) has yielded important insights into model behaviour, yet recurring pitfalls persist: findings that do not generalise, and causal interpretations that outrun the evidence. Our position is that causal inference specifies what constitutes a vali…