Explainable Artificial Intelligence (XAI)

Derniers papiers

• Why Linear Interpretability Works: Invariant Subspaces as a Result of Architectural Constraints

  Andres Saurez, Yousung Lee, Dongsoo Har · 11 février 2026

  Linear probes and sparse autoencoders consistently recover meaningful structure from transformer representations -- yet why should such simple methods succeed in deep, nonlinear systems? We show this is not merely an empirical regularity but a consequence of architectural necessity: transformers com…

• The effect of whitening on explanation performance

  Benedict Clark, Stoyan Karastoyanov, Rick Wilming, Stefan Haufe · 11 février 2026

  Explainable Artificial Intelligence (XAI) aims to provide transparent insights into machine learning models, yet the reliability of many feature attribution methods remains a critical challenge. Prior research (Haufe et al., 2014; Wilming et al., 2022, 2023) has demonstrated that these methods often…

• Behavioral Economics of AI: LLM Biases and Corrections

  Pietro Bini, Lin William Cong, Xing Huang, Lawrence J. Jin · 11 février 2026

  Do generative AI models, particularly large language models (LLMs), exhibit systematic behavioral biases in economic and financial decisions? If so, how can these biases be mitigated? Drawing on the cognitive psychology and experimental economics literatures, we conduct the most comprehensive set of…

• Seeing the Goal, Missing the Truth: Human Accountability for AI Bias

  Sean Cao, Wei Jiang, Hui Xu · 11 février 2026

  This research explores how human-defined goals influence the behavior of Large Language Models (LLMs) through purpose-conditioned cognition. Using financial prediction tasks, we show that revealing the downstream use (e.g., predicting stock returns or earnings) of LLM outputs leads the LLM to genera…

• Counterfactual Maps: What They Are and How to Find Them

  Awa Khouna, Julien Ferry, Thibaut Vidal · 11 février 2026

  Counterfactual explanations are a central tool in interpretable machine learning, yet computing them exactly for complex models remains challenging. For tree ensembles, predictions are piecewise constant over a large collection of axis-aligned hyperrectangles, implying that an optimal counterfactual…

• BiasScope: Towards Automated Detection of Bias in LLM-as-a-Judge Evaluation

  Peng Lai, Zhihao Ou, Yong Wang, Longyue Wang, Jian Yang, Yun Chen, Guanhua Chen · 11 février 2026

  LLM-as-a-Judge has been widely adopted across various research and practical applications, yet the robustness and reliability of its evaluation remain a critical issue. A core challenge it faces is bias, which has primarily been studied in terms of known biases and their impact on evaluation outcome…

• Feature salience -- not task-informativeness -- drives machine learning model explanations

  Benedict Clark, Marta Oliveira, Rick Wilming, Stefan Haufe · 11 février 2026

  Explainable AI (XAI) promises to provide insight into machine learning models' decision processes, where one goal is to identify failures such as shortcut learning. This promise relies on the field's assumption that input features marked as important by an XAI must contain information about the targ…

• CoMMa: Contribution-Aware Medical Multi-Agents From A Game-Theoretic Perspective

  Yichen Wu, Yujin Oh, Sangjoon Park, Kailong Fan, Dania Daye, Hana Farzaneh, Xiang Li, Raul Uppot, Quanzheng Li · 11 février 2026

  Recent multi-agent frameworks have broadened the ability to tackle oncology decision support tasks that require reasoning over dynamic, heterogeneous patient data. We propose Contribution-Aware Medical Multi-Agents (CoMMa), a decentralized LLM-agent framework in which specialists operate on partitio…

• Computing Conditional Shapley Values Using Tabular Foundation Models

  Lars Henry Berge Olsen, Dennis Christensen · 11 février 2026

  Shapley values have become a cornerstone of explainable AI, but they are computationally expensive to use, especially when features are dependent. Evaluating them requires approximating a large number of conditional expectations, either via Monte Carlo integration or regression. Until recently it ha…

• Generative AI Adoption in an Energy Company: Exploring Challenges and Use Cases

  Malik Abdul Sami, Zeeshan Rasheed, Meri Olenius, Muhammad Waseem, Kai-Kristian Kemell, Jussi Rasku, Pekka Abrahamsson · 11 février 2026

  Organisations are examining how generative AI can support their operational work and decision-making processes. This study investigates how employees in a energy company understand AI adoption and identify areas where AI and LLMs-based agentic workflows could assist daily activities. Data was collec…

• Features as Rewards: Scalable Supervision for Open-Ended Tasks via Interpretability

  Aaditya Vikram Prasad, Connor Watts, Jack Merullo, Dhruvil Gala, Owen Lewis, Thomas McGrath, Ekdeep Singh Lubana · 11 février 2026

  Language models trained on large-scale datasets have been shown to learn features that encode abstract concepts such as factuality or intent. Such features are traditionally used for test-time monitoring or steering. We present an alternative affordance: features as scalable supervision for open-end…

• Free(): Learning to Forget in Malloc-Only Reasoning Models

  Yilun Zheng, Dongyang Ma, Tian Liang, Jiahao Xu, Xinting Huang, Lijie Chen, Haitao Mi, Yan Wang · 10 février 2026

  Reasoning models enhance problem-solving by scaling test-time compute, yet they face a critical paradox: excessive thinking tokens often degrade performance rather than improve it. We attribute this to a fundamental architectural flaw: standard LLMs operate as "malloc-only" engines, continuously acc…

• Quantifying Explanation Quality in Graph Neural Networks using Out-of-Distribution Generalization

  Ding Zhang, Siddharth Betala, Chirag Agarwal · 10 février 2026

  Evaluating the quality of post-hoc explanations for Graph Neural Networks (GNNs) remains a significant challenge. While recent years have seen an increasing development of explainability methods, current evaluation metrics (e.g., fidelity, sparsity) often fail to assess whether an explanation identi…

• Structure-Aware Robust Counterfactual Explanations via Conditional Gaussian Network Classifiers

  Zhan-Yi Liao, Jaewon Yoo, Hao-Tsung Yang, Po-An Chen · 10 février 2026

  Counterfactual explanation (CE) is a core technique in explainable artificial intelligence (XAI), widely used to interpret model decisions and suggest actionable alternatives. This work presents a structure-aware and robustness-oriented counterfactual search method based on the conditional Gaussian …

• Learning to Judge: LLMs Designing and Applying Evaluation Rubrics

  Clemencia Siro, Pourya Aliannejadi, Mohammad Aliannejadi · 10 février 2026

  Large language models (LLMs) are increasingly used as evaluators for natural language generation, applying human-defined rubrics to assess system outputs. However, human rubrics are often static and misaligned with how models internally represent language quality. We introduce GER-Eval (Generating E…

• Circuit Representations of Random Forests with Applications to XAI

  Chunxi Ji, Adnan Darwiche · 10 février 2026

  We make three contributions in this paper. First, we present an approach for compiling a random forest classifier into a set of circuits, where each circuit directly encodes the instances in some class of the classifier. We show empirically that our proposed approach is significantly more efficient …

• An Information-Theoretic Framework for Comparing Voice and Text Explainability

  Mona Rajhans, Vishal Khawarey · 10 février 2026

  Explainable Artificial Intelligence (XAI) aims to make machine learning models transparent and trustworthy, yet most current approaches communicate explanations visually or through text. This paper introduces an information theoretic framework for analyzing how explanation modality specifically, voi…

• An Explainable Multi-Task Similarity Measure: Integrating Accumulated Local Effects and Weighted Fr\'echet Distance

  Pablo Hidalgo, Daniel Rodriguez · 10 février 2026

  In many machine learning contexts, tasks are often treated as interconnected components with the goal of leveraging knowledge transfer between them, which is the central aim of Multi-Task Learning (MTL). Consequently, this multi-task scenario requires addressing critical questions: which tasks are s…

• Noise Stability of Transformer Models

  Themistoklis Haris, Zihan Zhang, Yuichi Yoshida · 10 février 2026

  Understanding simplicity biases in deep learning offers a promising path toward developing reliable AI. A common metric for this, inspired by Boolean function analysis, is average sensitivity, which captures a model's robustness to single-token perturbations. We argue that average sensitivity has tw…

• Rethinking Explainable Disease Prediction: Synergizing Accuracy and Reliability via Reflective Cognitive Architecture

  Zijian Shao, Haiyang Shen, Mugeng Liu, Gecheng Fu, Yaoqi Guo, Yanfeng Wang, Yun Ma · 10 février 2026

  In clinical decision-making, predictive models face a persistent trade-off: accurate models are often opaque "black boxes," while interpretable methods frequently lack predictive precision or statistical grounding. In this paper, we challenge this dichotomy, positing that high predictive accuracy an…

• Emergent Structured Representations Support Flexible In-Context Inference in Large Language Models

  Ningyu Xu, Qi Zhang, Xipeng Qiu, Xuanjing Huang · 10 février 2026

  Large language models (LLMs) exhibit emergent behaviors suggestive of human-like reasoning. While recent work has identified structured, human-like conceptual representations within these models, it remains unclear whether they functionally rely on such representations for reasoning. Here we investi…

• Prospect Theory Fails for LLMs: Revealing Instability of Decision-Making under Epistemic Uncertainty

  Rui Wang, Qihan Lin, Jiayu Liu, Qing Zong, Tianshi Zheng, Weiqi Wang, Yangqiu Song · 10 février 2026

  Prospect Theory (PT) models human decision-making under uncertainty, while epistemic markers (e.g., maybe) serve to express uncertainty in language. However, it remains largely unexplored whether Prospect Theory applies to contemporary Large Language Models and whether epistemic markers, which expre…

• An Explainable Multi-Task Similarity Measure: Integrating Accumulated Local Effects and Weighted Fr\'echet Distance

  Pablo Hidalgo, Daniel Rodriguez · 10 février 2026

  In many machine learning contexts, tasks are often treated as interconnected components with the goal of leveraging knowledge transfer between them, which is the central aim of Multi-Task Learning (MTL). Consequently, this multi-task scenario requires addressing critical questions: which tasks are s…

• Data-Aware and Scalable Sensitivity Analysis for Decision Tree Ensembles

  Namrita Varshney, Ashutosh Gupta, Arhaan Ahmad, Tanay V. Tayal, S. Akshay · 10 février 2026

  Decision tree ensembles are widely used in critical domains, making robustness and sensitivity analysis essential to their trustworthiness. We study the feature sensitivity problem, which asks whether an ensemble is sensitive to a specified subset of features -- such as protected attributes -- whose…

• Where Not to Learn: Prior-Aligned Training with Subset-based Attribution Constraints for Reliable Decision-Making

  Ruoyu Chen, Shangquan Sun, Xiaoqing Guo, Sanyi Zhang, Kangwei Liu, Shiming Liu, Zhangcheng Wang, Qunli Zhang, Hua Zhang, Xiaochun Cao · 10 février 2026

  Reliable models should not only predict correctly, but also justify decisions with acceptable evidence. Yet conventional supervised learning typically provides only class-level labels, allowing models to achieve high accuracy through shortcut correlations rather than the intended evidence. Human pri…