Explainable Artificial Intelligence (XAI)

Derniers papiers

• Experiments or Outcomes? Probing Scientific Feasibility in Large Language Models

  Seyedali Mohammadi, Manas Gaur, Francis Ferraro · 22 avril 2026

  Scientific feasibility assessment asks whether a claim is consistent with established knowledge and whether experimental evidence could support or refute it. We frame feasibility assessment as a diagnostic reasoning task in which, given a hypothesis, a model predicts feasible or infeasible and justi…

• Four-Axis Decision Alignment for Long-Horizon Enterprise AI Agents

  Vasundra Srininvasan · 22 avril 2026

  Long-horizon enterprise agents make high-stakes decisions (loan underwriting, claims adjudication, clinical review, prior authorization) under lossy memory, multi-step reasoning, and binding regulatory constraints. Current evaluation reports a single task-success scalar that conflates distinct failu…

• Right for the Wrong Reasons: Epistemic Regret Minimization for LLM Causal Reasoning

  Edward Y. Chang · 22 avril 2026

  Large language models may answer causal questions correctly for the wrong reasons, substituting associational shortcuts P(Y|X) for the interventional query P(Y|do(X)). Current RL methods reward what the model answers but not why, reinforcing these shortcuts until distribution shift exposes them. We …

• Beyond Coefficients: Forecast-Necessity Testing for Interpretable Causal Discovery in Nonlinear Time-Series Models

  Valentina Kuskova, Dmitry Zaytsev, Michael Coppedge · 22 avril 2026

  Nonlinear machine-learning models are increasingly used to discover causal relationships in time-series data, yet the interpretation of their outputs remains poorly understood. In particular, causal scores produced by regularized neural autoregressive models are often treated as analogues of regress…

• TACENR: Task-Agnostic Contrastive Explanations for Node Representations

  Vasiliki Papanikou, Evaggelia Pitoura · 22 avril 2026

  Graph representation learning has achieved notable success in encoding graph-structured data into latent vector spaces, enabling a wide range of downstream tasks. However, these node representations remain opaque and difficult to interpret. Existing explainability methods primarily focus on supervis…

• PREF-XAI: Preference-Based Personalized Rule Explanations of Black-Box Machine Learning Models

  Salvatore Greco, Jacek Karolczak, Roman S{\l}owi\'nski, Jerzy Stefanowski · 22 avril 2026

  Explainable artificial intelligence (XAI) has predominantly focused on generating model-centric explanations that approximate the behavior of black-box models. However, such explanations often overlook a fundamental aspect of interpretability: different users require different explanations depending…

• ParamBoost: Gradient Boosted Piecewise Cubic Polynomials

  Nicolas Salvad\'e, Tim Hillel · 22 avril 2026

  Generalized Additive Models (GAMs) can be used to create non-linear glass-box (i.e. explicitly interpretable) models, where the predictive function is fully observable over the complete input space. However, glass-box interpretability itself does not allow for the incorporation of expert knowledge f…

• From Top-1 to Top-K: A Reproducibility Study and Benchmarking of Counterfactual Explanations for Recommender Systems

  Quang-Huy Nguyen, Thanh-Hai Nguyen, Khac-Manh Thai, Duc-Hoang Pham, Huy-Son Nguyen, Cam-Van Thi Nguyen, Masoud Mansoury, Duc-Trong Le, Hoang-Quynh Le · 22 avril 2026

  Counterfactual explanations (CEs) provide an intuitive way to understand recommender systems by identifying minimal modifications to user-item interactions that alter recommendation outcomes. Existing CE methods for recommender systems, however, have been evaluated under heterogeneous protocols, usi…

• Symbolic Quantile Regression for the Interpretable Prediction of Conditional Quantiles

  Cas Oude Hoekstra, Floris den Hengst · 22 avril 2026

  Symbolic Regression (SR) is a well-established framework for generating interpretable or white-box predictive models. Although SR has been successfully applied to create interpretable estimates of the average of the outcome, it is currently not well understood how it can be used to estimate the rela…

• Where Fake Citations Are Made: Tracing Field-Level Hallucination to Specific Neurons in LLMs

  Yuefei Chen, Yihao Quan, Xiaodong Lin, Ruixiang Tang · 22 avril 2026

  LLMs frequently generate fictitious yet convincing citations, often expressing high confidence even when the underlying reference is wrong. We study this failure across 9 models and 108{,}000 generated references, and find that author names fail far more often than other fields across all models and…

• TEMPO: Scaling Test-time Training for Large Reasoning Models

  Qingyang Zhang, Xinke Kong, Haitao Wu, Qinghua Hu, Minghao Wu, Baosong Yang, Yu Cheng, Yun Luo, Ganqu Cui, Changqing Zhang · 22 avril 2026

  Test-time training (TTT) adapts model parameters on unlabeled test instances during inference time, which continuously extends capabilities beyond the reach of offline training. Despite initial gains, existing TTT methods for LRMs plateau quickly and do not benefit from additional test-time compute.…

• Easy Samples Are All You Need: Self-Evolving LLMs via Data-Efficient Reinforcement Learning

  Zhiyin Yu, Bo Zhang, Qibin Hou, Zhonghai Wu, Xiao Luo, Lei Bai · 22 avril 2026

  Previous LLMs-based RL studies typically follow either supervised learning with high annotation costs, or unsupervised paradigms using voting or entropy-based rewards. However, their performance remains far from satisfactory due to the substantial annotation cost and issues such as model collapse or…

• FSEVAL: Feature Selection Evaluation Toolbox and Dashboard

  Muhammad Rajabinasab, Arthur Zimek · 21 avril 2026

  Feature selection is a fundamental machine learning and data mining task, involved with discriminating redundant features from informative ones. It is an attempt to address the curse of dimensionality by removing the redundant features, while unlike dimensionality reduction methods, preserving expla…

• Functional Similarity Metric for Neural Networks: Overcoming Parametric Ambiguity via Activation Region Analysis

  Kutomanov Hennadii · 21 avril 2026

  As modern deep learning architectures grow in complexity, representational ambiguity emerges as a critical barrier to their interpretability and reliable merging. For ReLU networks, identical functional mappings can be achieved through entirely different weight configurations due to algebraic symmet…

• A Probabilistic Consensus-Driven Approach for Robust Counterfactual Explanations

  Marcin Kostrzewa, Maciej Zi\k{e}ba, Jerzy Stefanowski · 21 avril 2026

  Counterfactual explanations (CFEs) are essential for interpreting black-box models, yet they often become invalid when models are slightly changed. Existing methods for generating robust CFEs are often limited to specific types of models, require costly tuning, or inflexible robustness controls. We …

• ExAI5G: A Logic-Based Explainable AI Framework for Intrusion Detection in 5G Networks

  Saeid Sheikhi, Panos Kostakos, Lauri Loven · 21 avril 2026

  Intrusion detection systems (IDSs) for 5G networks must handle complex, high-volume traffic. Although opaque "black-box" models can achieve high accuracy, their lack of transparency hinders trust and effective operational response. We propose ExAI5G, a framework that prioritizes interpretability by …

• Annotation Entropy Predicts Per-Example Learning Dynamics in LoRA Fine-Tuning

  Brady Steele · 21 avril 2026

  We find that LoRA fine-tuning exhibits un-learning on contested examples: items with high annotator disagreement show increasing loss during training, a qualitatively distinct pattern largely absent under full fine-tuning and consistent across all six models tested (four encoder, two decoder-only). …

• A Sugeno Integral View of Binarized Neural Network Inference

  Isma\"il Baaj, Henri Prade · 21 avril 2026

  In this article, we establish a precise connection between binarized neural networks (BNNs) and Sugeno integrals. The advantage of the Sugeno integral is that it provides a framework for representing the importance of inputs and their interactions, while being equivalent to a set of if-then rules. F…

• Agents Explore but Agents Ignore: LLMs Lack Environmental Curiosity

  Leon Engl\"ander, Sophia Althammer, Ahmet \"Ust\"un, Matthias Gall\'e, Tom Sherborne · 21 avril 2026

  LLM-based agents are assumed to integrate environmental observations into their reasoning: discovering highly relevant but unexpected information should naturally lead to a model exploiting its own discoveries. We show that this assumption is false for current LLM-based agents, which struggle to ref…

• ProtoTTA: Prototype-Guided Test-Time Adaptation

  Mohammad Mahdi Abootorabi, Parvin Mousavi, Purang Abolmaesumi, Evan Shelhamer · 20 avril 2026

  Deep networks that rely on prototypes-interpretable representations that can be related to the model input-have gained significant attention for balancing high accuracy with inherent interpretability, which makes them suitable for critical domains such as healthcare. However, these models are limite…

• The World Leaks the Future: Harness Evolution for Future Prediction Agents

  Chuyang Wei (University of Science and Technology of China, Zhongguancun Academy, Beijing, China), Maohang Gao (University of Science and Technology of China, Zhongguancun Academy, Beijing, China), Zhixin Han (Zhongguancun Academy, Beijing, China), Kefei Chen (Zhongguancun Academy, Beijing, China, Tsinghua University), Yu Zhuang (Zhongguancun Academy, Beijing, China), Haoxiang Guan (University of Science and Technology of China, Zhongguancun Academy, Beijing, China), Yanzhi Zhang (Zhongguancun Academy, Beijing, China), Yilin Cheng (Zhongguancun Academy, Beijing, China), Jiyan He (Zhongguancun Academy, Beijing, China), Huanhuan Chen (University of Science and Technology of China), Jian Li (Tsinghua University), Yu Shi (Zhongguancun Academy, Beijing, China), Yitong Duan (Zhongguancun Academy, Beijing, China), Shuxin Zheng (Zhongguancun Academy, Beijing, China) · 20 avril 2026

  Many consequential decisions must be made before the relevant outcome is known. Such problems are commonly framed as \emph{future prediction}, where an LLM agent must form a prediction for an unresolved question using only the public information available at the prediction time. The setting is diffi…

• Towards Rigorous Explainability by Feature Attribution

  Olivier L\'etoff\'e, Xuanxiang Huang, Joao Marques-Silva · 20 avril 2026

  For around a decade, non-symbolic methods have been the option of choice when explaining complex machine learning (ML) models. Unfortunately, such methods lack rigor and can mislead human decision-makers. In high-stakes uses of ML, the lack of rigor is especially problematic. One prime example of pr…

• Applied Explainability for Large Language Models: A Comparative Study

  Venkata Abhinandan Kancharla · 20 avril 2026

  Large language models (LLMs) achieve strong performance across many natural language processing tasks, yet their decision processes remain difficult to interpret. This lack of transparency creates challenges for trust, debugging, and deployment in real-world systems. This paper presents an applied…

• AtManRL: Towards Faithful Reasoning via Differentiable Attention Saliency

  Max Henning H\"oth, Kristian Kersting, Bj\"orn Deiseroth, Letitia Parcalabescu · 20 avril 2026

  Large language models (LLMs) increasingly rely on chain-of-thought (CoT) reasoning to solve complex tasks. Yet ensuring that the reasoning trace both contributes to and faithfully reflects the processes underlying the model's final answer, rather than merely accompanying it, remains challenging. We …

• Token Statistics Reveal Conversational Drift in Multi-turn LLM Interaction

  Wael Hafez, Amir Nazeri · 20 avril 2026

  Large language models, LLMs, are increasingly deployed in multiturn settings where earlier responses shape later ones, making reliability dependent on whether a conversation remains consistent over time. When this consistency degrades undetected, downstream decisions lose their grounding in the exch…