Explainable Artificial Intelligence (XAI)

Derniers papiers

• Soft decision trees for survival analysis

  Antonio Consolo, Edoardo Amaldi, Emilio Carrizosa · 24 novembre 2025

  Decision trees are popular in survival analysis for their interpretability and ability to model complex relationships. Survival trees, which predict the timing of singular events using censored historical data, are typically built through heuristic approaches. Recently, there has been growing intere…

• Deep Improvement Supervision

  Arip Asadulaev, Rayan Banerjee, Fakhri Karray, Martin Takac · 24 novembre 2025

  Recently, it was shown that small, looped architectures, such as Tiny Recursive Models (TRMs), can outperform Large Language Models (LLMs) on complex reasoning tasks, including the Abstraction and Reasoning Corpus (ARC). In this work, we investigate a core question: how can we further improve the ef…

• Structured Debate Improves Corporate Credit Reasoning in Financial AI

  Yoonjin Lee, Munhee Kim, Hanbi Choi, Juhyeon Park, Seungho Lyoo, Woojin Park · 24 novembre 2025

  Despite advances in financial AI, the automation of evidence-based reasoning remains unresolved in corporate credit assessment, where qualitative non-financial indicators exert decisive influence on loan repayment outcomes yet resist formalization. Existing approaches focus predominantly on numerica…

• Comprehensive Evaluation of Prototype Neural Networks

  Philipp Schlinge, Steffen Meinert, Martin Atzmueller · 24 novembre 2025

  Prototype models are an important method for explainable artificial intelligence (XAI) and interpretable machine learning. In this paper, we perform an in-depth analysis of a set of prominent prototype models including ProtoPNet, ProtoPool and PIPNet. For their assessment, we apply a comprehensive s…

• Pass@k Metric for RLVR: A Diagnostic Tool of Exploration, But Not an Objective

  Yang Yu · 21 novembre 2025

  The ability of Large Language Models (LLMs) to perform complex, multi-step reasoning is a central focus of modern AI research. To evaluate and enhance this capability, the pass@k metric, which measures the probability of obtaining at least one correct solution in k independent samples, has received …

• When concept-based XAI is imprecise: Do people distinguish between generalisations and misrepresentations?

  Romy M\"uller · 21 novembre 2025

  Concept-based explainable artificial intelligence (C-XAI) can let people see which representations an AI model has learned. This is particularly important when high-level semantic information (e.g., actions and relations) is used to make decisions about abstract categories (e.g., danger). In such ta…

• Formal Abductive Latent Explanations for Prototype-Based Networks

  Jules Soria, Zakaria Chihani, Julien Girard-Satabin, Alban Grastien, Romain Xu-Darme, Daniela Cancila · 21 novembre 2025

  Case-based reasoning networks are machine-learning models that make predictions based on similarity between the input and prototypical parts of training samples, called prototypes. Such models are able to explain each decision by pointing to the prototypes that contributed the most to the final outc…

• iLTM: Integrated Large Tabular Model

  David Bonet, Mar\c{c}al Comajoan Cara, Alvaro Calafell, Daniel Mas Montserrat, Alexander G. Ioannidis · 21 novembre 2025

  Tabular data underpins decisions across science, industry, and public services. Despite rapid progress, advances in deep learning have not fully carried over to the tabular domain, where gradient-boosted decision trees (GBDTs) remain a default choice in practice. We present iLTM, an integrated Large…

• The Oracle and The Prism: A Decoupled and Efficient Framework for Generative Recommendation Explanation

  Jiaheng Zhang, Daqiang Zhang · 21 novembre 2025

  The integration of Large Language Models (LLMs) into explainable recommendation systems often leads to a performance-efficiency trade-off in end-to-end architectures, where joint optimization of ranking and explanation can result in suboptimal compromises. To resolve this, we propose Prism, a novel …

• Learning from Sufficient Rationales: Analysing the Relationship Between Explanation Faithfulness and Token-level Regularisation Strategies

  Jonathan Kamp, Lisa Beinborn, Antske Fokkens · 21 novembre 2025

  Human explanations of natural language, rationales, form a tool to assess whether models learn a label for the right reasons or rely on dataset-specific shortcuts. Sufficiency is a common metric for estimating the informativeness of rationales, but it provides limited insight into the effects of rat…

• Correlation-Aware Feature Attribution Based Explainable AI

  Poushali Sengupta, Yan Zhang, Frank Eliassen, Sabita Maharjan · 21 novembre 2025

  Explainable AI (XAI) is increasingly essential as modern models become more complex and high-stakes applications demand transparency, trust, and regulatory compliance. Existing global attribution methods often incur high computational costs, lack stability under correlated inputs, and fail to scale …

• From Performance to Understanding: A Vision for Explainable Automated Algorithm Design

  Niki van Stein, Anna V. Kononova, Thomas B\"ack · 21 novembre 2025

  Automated algorithm design is entering a new phase: Large Language Models can now generate full optimisation (meta)heuristics, explore vast design spaces and adapt through iterative feedback. Yet this rapid progress is largely performance-driven and opaque. Current LLM-based approaches rarely reveal…

• MACIE: Multi-Agent Causal Intelligence Explainer for Collective Behavior Understanding

  Abraham Itzhak Weinberg · 21 novembre 2025

  As Multi Agent Reinforcement Learning systems are used in safety critical applications. Understanding why agents make decisions and how they achieve collective behavior is crucial. Existing explainable AI methods struggle in multi agent settings. They fail to attribute collective outcomes to individ…

• Fast-DataShapley: Neural Modeling for Training Data Valuation

  Haifeng Sun, Yu Xiong, Runze Wu, Xinyu Cai, Changjie Fan, Lan Zhang, Xiang-Yang Li · 21 novembre 2025

  The value and copyright of training data are crucial in the artificial intelligence industry. Service platforms should protect data providers' legitimate rights and fairly reward them for their contributions. Shapley value, a potent tool for evaluating contributions, outperforms other methods in the…

• Towards Understanding Layer Contributions in Tabular In-Context Learning Models

  Amir Rezaei Balef, Mykhailo Koshil, Katharina Eggensperger · 20 novembre 2025

  Despite the architectural similarities between tabular in-context learning (ICL) models and large language models (LLMs), little is known about how individual layers contribute to tabular prediction. In this paper, we investigate how the latent spaces evolve across layers in tabular ICL models, iden…

• A Unified Framework for Provably Efficient Algorithms to Estimate Shapley Values

  Tyler Chen, Akshay Seshadri, Mattia J. Villani, Pradeep Niroula, Shouvanik Chakrabarti, Archan Ray, Pranav Deshpande, Romina Yalovetzky, Marco Pistoia, Niraj Kumar · 20 novembre 2025

  Shapley values have emerged as a critical tool for explaining which features impact the decisions made by machine learning models. However, computing exact Shapley values is difficult, generally requiring an exponential (in the feature dimension) number of model evaluations. To address this, many mo…

• CID: Measuring Feature Importance Through Counterfactual Distributions

  Eddie Conti, \'Alvaro Parafita, Axel Brando · 20 novembre 2025

  Assessing the importance of individual features in Machine Learning is critical to understand the model's decision-making process. While numerous methods exist, the lack of a definitive ground truth for comparison highlights the need for alternative, well-founded measures. This paper introduces a no…

• Soft-Label Training Preserves Epistemic Uncertainty

  Agamdeep Singh, Ashish Tiwari, Hosein Hasanbeig, Priyanshu Gupta · 19 novembre 2025

  Many machine learning tasks involve inherent subjectivity, where annotators naturally provide varied labels. Standard practice collapses these label distributions into single labels, aggregating diverse human judgments into point estimates. We argue that this approach is epistemically misaligned for…

• Data Whitening Improves Sparse Autoencoder Learning

  Ashwin Saraswatula, David Klindt · 19 novembre 2025

  Sparse autoencoders (SAEs) have emerged as a promising approach for learning interpretable features from neural network activations. However, the optimization landscape for SAE training can be challenging due to correlations in the input data. We demonstrate that applying PCA Whitening to input acti…

• EchoAgent: Guideline-Centric Reasoning Agent for Echocardiography Measurement and Interpretation

  Matin Daghyani, Lyuyang Wang, Nima Hashemi, Bassant Medhat, Baraa Abdelsamad, Eros Rojas Velez, XiaoXiao Li, Michael Y. C. Tsang, Christina Luong, Teresa S. M. Tsang, Purang Abolmaesumi · 19 novembre 2025

  Purpose: Echocardiographic interpretation requires video-level reasoning and guideline-based measurement analysis, which current deep learning models for cardiac ultrasound do not support. We present EchoAgent, a framework that enables structured, interpretable automation for this domain. Methods: E…

• Explaining Similarity in Vision-Language Encoders with Weighted Banzhaf Interactions

  Hubert Baniecki, Maximilian Muschalik, Fabian Fumagalli, Barbara Hammer, Eyke H\"ullermeier, Przemyslaw Biecek · 19 novembre 2025

  Language-image pre-training (LIP) enables the development of vision-language models capable of zero-shot classification, localization, multimodal retrieval, and semantic understanding. Various explanation methods have been proposed to visualize the importance of input image-text pairs on the model's…

• ScoresActivation: A New Activation Function for Model Agnostic Global Explainability by Design

  Emanuel Covaci, Fabian Galis, Radu Balan, Daniela Zaharie, Darian Onchis · 19 novembre 2025

  Understanding the decision of large deep learning models is a critical challenge for building transparent and trustworthy systems. Although the current post hoc explanation methods offer valuable insights into feature importance, they are inherently disconnected from the model training process, limi…

• Consistency-based Abductive Reasoning over Perceptual Errors of Multiple Pre-trained Models in Novel Environments

  Mario Leiva, Noel Ngu, Joshua Shay Kricheli, Aditya Taparia, Ransalu Senanayake, Paulo Shakarian, Nathaniel Bastian, John Corcoran, Gerardo Simari · 18 novembre 2025

  The deployment of pre-trained perception models in novel environments often leads to performance degradation due to distributional shifts. Although recent artificial intelligence approaches for metacognition use logical rules to characterize and filter model errors, improving precision often comes a…

• Not All Attention Heads Are What You Need: Refining CLIP's Image Representation with Attention Ablation

  Feng Lin, Marco Chen, Haokui Zhang, Xiaotian Yu, Guangming Lu, Rong Xiao · 18 novembre 2025

  This paper investigates the role of attention heads in CLIP's image encoder. Building on interpretability studies, we conduct an exhaustive analysis and find that certain heads, distributed across layers, are detrimental to the resulting representations. To mitigate their impact, we propose a simple…

• DeceptionBench: A Comprehensive Benchmark for AI Deception Behaviors in Real-world Scenarios

  Yao Huang, Yitong Sun, Yichi Zhang, Ruochen Zhang, Yinpeng Dong, Xingxing Wei · 18 novembre 2025

  Despite the remarkable advances of Large Language Models (LLMs) across diverse cognitive tasks, the rapid enhancement of these capabilities also introduces emergent deceptive behaviors that may induce severe risks in high-stakes deployments. More critically, the characterization of deception across …