Explainable Artificial Intelligence (XAI)

Derniers papiers

• Conditional Coverage Diagnostics for Conformal Prediction

  Sacha Braun, David Holzm\"uller, Michael I. Jordan, Francis Bach · 15 décembre 2025

  Evaluating conditional coverage remains one of the most persistent challenges in assessing the reliability of predictive systems. Although conformal methods can give guarantees on marginal coverage, no method can guarantee to produce sets with correct conditional coverage, leaving practitioners with…

• Visualizing token importance for black-box language models

  Paulius Rauba, Qiyao Wei, Mihaela van der Schaar · 15 décembre 2025

  We consider the problem of auditing black-box large language models (LLMs) to ensure they behave reliably when deployed in production settings, particularly in high-stakes domains such as legal, medical, and regulatory compliance. Existing approaches for LLM auditing often focus on isolated aspects …

• EmeraldMind: A Knowledge Graph-Augmented Framework for Greenwashing Detection

  Georgios Kaoukis, Ioannis Aris Koufopoulos, Psaroudaki Eleni, Danae Pla Karidi, Evaggelia Pitoura, George Papastefanatos, Panayiotis Tsaparas · 15 décembre 2025

  As AI and web agents become pervasive in decision-making, it is critical to design intelligent systems that not only support sustainability efforts but also guard against misinformation. Greenwashing, i.e., misleading corporate sustainability claims, poses a major challenge to environmental progress…

• Provable Recovery of Locally Important Signed Features and Interactions from Random Forest

  Kata Vuk, Nicolas Alexander Ihlo, Merle Behr · 15 décembre 2025

  Feature and Interaction Importance (FII) methods are essential in supervised learning for assessing the relevance of input variables and their interactions in complex prediction models. In many domains, such as personalized medicine, local interpretations for individual predictions are often require…

• Cognitive Mirrors: Exploring the Diverse Functional Roles of Attention Heads in LLM Reasoning

  Xueqi Ma, Jun Wang, Yanbei Jiang, Sarah Monazam Erfani, Tongliang Liu, James Bailey · 15 décembre 2025

  Large language models (LLMs) have achieved state-of-the-art performance in a variety of tasks, but remain largely opaque in terms of their internal mechanisms. Understanding these mechanisms is crucial to improve their reasoning abilities. Drawing inspiration from the interplay between neural proces…

• CIP: A Plug-and-Play Causal Prompting Framework for Mitigating Hallucinations under Long-Context Noise

  Qingsen Ma, Dianyun Wang, Ran Jing, Yujun Sun, Zhenbo Xu · 15 décembre 2025

  Large language models often hallucinate when processing long and noisy retrieval contexts because they rely on spurious correlations rather than genuine causal relationships. We propose CIP, a lightweight and plug-and-play causal prompting framework that mitigates hallucinations at the input stage. …

• Explanation Bias is a Product: Revealing the Hidden Lexical and Position Preferences in Post-Hoc Feature Attribution

  Jonathan Kamp, Roos Bakker, Dominique Blok · 15 décembre 2025

  Good quality explanations strengthen the understanding of language models and data. Feature attribution methods, such as Integrated Gradient, are a type of post-hoc explainer that can provide token-level insights. However, explanations on the same input may vary greatly due to underlying biases of d…

• Back to the Baseline: Examining Baseline Effects on Explainability Metrics

  Agustin Martin Picard (ANITI), Thibaut Boissin (ANITI), Varshini Subhash (SU), R\'emi Cad\`ene (SU), Thomas Fel (ANITI) · 15 décembre 2025

  Attribution methods are among the most prevalent techniques in Explainable Artificial Intelligence (XAI) and are usually evaluated and compared using Fidelity metrics, with Insertion and Deletion being the most popular. These metrics rely on a baseline function to alter the pixels of the input image…

• Interpretable and Steerable Concept Bottleneck Sparse Autoencoders

  Akshay Kulkarni, Tsui-Wei Weng, Vivek Narayanaswamy, Shusen Liu, Wesam A. Sakla, Kowshik Thopalli · 12 décembre 2025

  Sparse autoencoders (SAEs) promise a unified approach for mechanistic interpretability, concept discovery, and model steering in LLMs and LVLMs. However, realizing this potential requires that the learned features be both interpretable and steerable. To that end, we introduce two new computationally…

• If generative AI is the answer, what is the question?

  Ambuj Tewari · 12 décembre 2025

  Beginning with text and images, generative AI has expanded to audio, video, computer code, and molecules. Yet, if generative AI is the answer, what is the question? We explore the foundations of generation as a distinct machine learning task with connections to prediction, compression, and decision-…

• Achieving Trustworthy Real-Time Decision Support Systems with Low-Latency Interpretable AI Models

  Zechun Deng, Ziwei Liu, Ziqian Bi, Junhao Song, Chia Xin Liang, Joe Yeong, Xinyuan Song, Junfeng Hao · 12 décembre 2025

  This paper investigates real-time decision support systems that leverage low-latency AI models, bringing together recent progress in holistic AI-driven decision tools, integration with Edge-IoT technologies, and approaches for effective human-AI teamwork. It looks into how large language models can …

• Classifier Reconstruction Through Counterfactual-Aware Wasserstein Prototypes

  Xuan Zhao, Zhuo Cao, Arya Bangun, Hanno Scharr, Ira Assent · 12 décembre 2025

  Counterfactual explanations provide actionable insights by identifying minimal input changes required to achieve a desired model prediction. Beyond their interpretability benefits, counterfactuals can also be leveraged for model reconstruction, where a surrogate model is trained to replicate the beh…

• Causal Reasoning Favors Encoders: On The Limits of Decoder-Only Models

  Amartya Roy, Elamparithy M, Kripabandhu Ghosh, Ponnurangam Kumaraguru, Adrian de Wynter · 12 décembre 2025

  In context learning (ICL) underpins recent advances in large language models (LLMs), although its role and performance in causal reasoning remains unclear. Causal reasoning demands multihop composition and strict conjunctive control, and reliance on spurious lexical relations of the input could prov…

• MedXAI: A Retrieval-Augmented and Self-Verifying Framework for Knowledge-Guided Medical Image Analysis

  Midhat Urooj, Ayan Banerjee, Farhat Shaikh, Kuntal Thakur, Sandeep Gupta · 12 décembre 2025

  Accurate and interpretable image-based diagnosis remains a fundamental challenge in medical AI, particularly un- der domain shifts and rare-class conditions. Deep learning mod- els often struggle with real-world distribution changes, exhibit bias against infrequent pathologies, and lack the transpar…

• GAIR: GUI Automation via Information-Joint Reasoning and Group Reflection

  Zishu Wei, Qixiang Ma, Xavier Hu, Yuhang Liu, Hui Zang, Yudong Zhao, Tao Wang, Shengyu Zhang, Fei Wu · 11 décembre 2025

  Building AI systems for GUI automation task has attracted remarkable research efforts, where MLLMs are leveraged for processing user requirements and give operations. However, GUI automation includes a wide range of tasks, from document processing to online shopping, from CAD to video editing. Diver…

• Financial Instruction Following Evaluation (FIFE)

  Glenn Matlin, Siddharth, Anirudh JM, Aditya Shukla, Yahya Hassan, Sudheer Chava · 11 décembre 2025

  Language Models (LMs) struggle with complex, interdependent instructions, particularly in high-stakes domains like finance where precision is critical. We introduce FIFE, a novel, high-difficulty benchmark designed to assess LM instruction-following capabilities for financial analysis tasks. FIFE co…

• GS-KAN: Parameter-Efficient Kolmogorov-Arnold Networks via Sprecher-Type Shared Basis Functions

  Oscar Eliasson · 11 décembre 2025

  The Kolmogorov-Arnold representation theorem offers a theoretical alternative to Multi-Layer Perceptrons (MLPs) by placing learnable univariate functions on edges rather than nodes. While recent implementations such as Kolmogorov-Arnold Networks (KANs) demonstrate high approximation capabilities, th…

• STACHE: Local Black-Box Explanations for Reinforcement Learning Policies

  Andrew Elashkin, Orna Grumberg · 11 décembre 2025

  Reinforcement learning agents often behave unexpectedly in sparse-reward or safety-critical environments, creating a strong need for reliable debugging and verification tools. In this paper, we propose STACHE, a comprehensive framework for generating local, black-box explanations for an agent's spec…

• Supervised learning pays attention

  Erin Craig, Robert Tibshirani · 11 décembre 2025

  In-context learning with attention enables large neural networks to make context-specific predictions by selectively focusing on relevant examples. Here, we adapt this idea to supervised learning procedures such as lasso regression and gradient boosting, for tabular data. Our goals are to (1) flexib…

• DW-KNN: A Transparent Local Classifier Integrating Distance Consistency and Neighbor Reliability

  Kumarjit Pathak, Karthik K, Sachin Madan, Jitin Kapila · 11 décembre 2025

  K-Nearest Neighbors (KNN) is one of the most used ML classifiers. However, if we observe closely, standard distance-weighted KNN and relative variants assume all 'k' neighbors are equally reliable. In heterogeneous feature space, this becomes a limitation that hinders reliability in predicting true …

• Interpreto: An Explainability Library for Transformers

  Antonin Poch\'e, Thomas Mullor, Gabriele Sarti, Fr\'ed\'eric Boisnard, Corentin Friedrich, Charlotte Claye, Fran\c{c}ois Hoofd, Raphael Bernas, C\'eline Hudelot, Fanny Jourdan · 11 décembre 2025

  Interpreto is a Python library for post-hoc explainability of text HuggingFace models, from early BERT variants to LLMs. It provides two complementary families of methods: attributions and concept-based explanations. The library connects recent research to practical tooling for data scientists, aimi…

• Calibrated Trust in Dealing with LLM Hallucinations: A Qualitative Study

  Adrian Ryser, Florian Allwein, Tim Schlippe · 11 décembre 2025

  Hallucinations are outputs by Large Language Models (LLMs) that are factually incorrect yet appear plausible [1]. This paper investigates how such hallucinations influence users' trust in LLMs and users' interaction with LLMs. To explore this in everyday use, we conducted a qualitative study with 19…

• Balanced Accuracy: The Right Metric for Evaluating LLM Judges - Explained through Youden's J statistic

  Stephane Collot, Colin Fraser, Justin Zhao, William F. Shen, Timon Willi, Ilias Leontiadis · 10 décembre 2025

  Rigorous evaluation of large language models (LLMs) relies on comparing models by the prevalence of desirable or undesirable behaviors, such as task pass rates or policy violations. These prevalence estimates are produced by a classifier, either an LLM-as-a-judge or human annotators, making the choi…

• Jacobian Aligned Random Forests

  Sarwesh Rauniyar · 10 décembre 2025

  Axis-aligned decision trees are fast and stable but struggle on datasets with rotated or interaction-dependent decision boundaries, where informative splits require linear combinations of features rather than single-feature thresholds. Oblique forests address this with per-node hyperplane splits, bu…

• Uncertainty-Aware Subset Selection for Robust Visual Explainability under Distribution Shifts

  Madhav Gupta, Vishak Prasad C, Ganesh Ramakrishnan · 10 décembre 2025

  Subset selection-based methods are widely used to explain deep vision models: they attribute predictions by highlighting the most influential image regions and support object-level explanations. While these methods perform well in in-distribution (ID) settings, their behavior under out-of-distributi…