Explainable Artificial Intelligence (XAI)

Derniers papiers

• Swap-guided Preference Learning for Personalized Reinforcement Learning from Human Feedback

  Gihoon Kim, Euntai Kim · 16 mars 2026

  Reinforcement Learning from Human Feedback (RLHF) is a widely used approach to align large-scale AI systems with human values. However, RLHF typically assumes a single, universal reward, which overlooks diverse preferences and limits personalization. Variational Preference Learning (VPL) seeks to ad…

• AgentDrift: Unsafe Recommendation Drift Under Tool Corruption Hidden by Ranking Metrics in LLM Agents

  Zekun Wu, Adriano Koshiyama, Sahan Bulathwela, Maria Perez-Ortiz · 16 mars 2026

  Tool-augmented LLM agents increasingly serve as multi-turn advisors in high-stakes domains, yet their evaluation relies on ranking-quality metrics that measure what is recommended but not whether it is safe for the user. We introduce a paired-trajectory protocol that replays real financial dialogues…

• Do LLMs Share Human-Like Biases? Causal Reasoning Under Prior Knowledge, Irrelevant Context, and Varying Compute Budgets

  Hanna M. Dettki, Charley M. Wu, Bob Rehder · 16 mars 2026

  Large language models (LLMs) are increasingly used in domains where causal reasoning matters, yet it remains unclear whether their judgments reflect normative causal computation, human-like shortcuts, or brittle pattern matching. We benchmark 20+ LLMs against a matched human baseline on 11 causal ju…

• When LLM Judge Scores Look Good but Best-of-N Decisions Fail

  Eddie Landesberg · 16 mars 2026

  Large language models are often used as judges to score candidate responses, then validated with a single global metric such as correlation with reference labels. This can be misleading when the real deployment task is best-of-n selection within a prompt. In a 5,000-prompt best-of-4 benchmark from…

• HCP-DCNet: A Hierarchical Causal Primitive Dynamic Composition Network for Self-Improving Causal Understanding

  Ming Lei, Shufan Wu, Christophe Baehr · 16 mars 2026

  The ability to understand and reason about cause and effect -- encompassing interventions, counterfactuals, and underlying mechanisms -- is a cornerstone of robust artificial intelligence. While deep learning excels at pattern recognition, it fundamentally lacks a model of causality, making systems …

• Are General-Purpose Vision Models All We Need for 2D Medical Image Segmentation? A Cross-Dataset Empirical Study

  Vanessa Borst, Samuel Kounev · 16 mars 2026

  Medical image segmentation (MIS) is a fundamental component of computer-assisted diagnosis and clinical decision support systems. Over the past decade, numerous architectures specifically tailored to medical imaging have emerged to address domain-specific challenges such as low contrast, small anato…

• From Garbage to Gold: A Data-Architectural Theory of Predictive Robustness

  Terrence J. Lee-St. John, Jordan L. Lawson, Bartlomiej Piechowski-Jozwiak · 16 mars 2026

  Tabular machine learning presents a paradox: modern models achieve state-of-the-art performance using high-dimensional (high-D), collinear, error-prone data, defying the "Garbage In, Garbage Out" mantra. To help resolve this, we synthesize principles from Information Theory, Latent Factor Models, an…

• Explainable AI Using Inherently Interpretable Components for Wearable-based Health Monitoring

  Maurice Kuschel, Solveig Vieluf, Claus Reinsberger, Tobias Loddenkemper, Tanuj Hasija · 16 mars 2026

  The use of wearables in medicine and wellness, enabled by AI-based models, offers tremendous potential for real-time monitoring and interpretable event detection. Explainable AI (XAI) is required to assess what models have learned and build trust in model outputs, for patients, healthcare profession…

• Delta1 with LLM: symbolic and neural integration for credible and explainable reasoning

  Yang Xu, Jun Liu, Shuwei Chen, Chris Nugent, Hailing Guo · 16 mars 2026

  Neuro-symbolic reasoning increasingly demands frameworks that unite the formal rigor of logic with the interpretability of large language models (LLMs). We introduce an end to end explainability by construction pipeline integrating the Automated Theorem Generator Delta1 based on the full triangular …

• Causality Is Key to Understand and Balance Multiple Goals in Trustworthy ML and Foundation Models

  Ruta Binkyte, Ivaxi Sheth, Zhijing Jin, Mohammad Havaei, Bernhard Sch\"olkopf, Mario Fritz · 16 mars 2026

  Ensuring trustworthiness in machine learning (ML) systems is crucial as they become increasingly embedded in high-stakes domains. This paper advocates for integrating causal methods into machine learning to navigate the trade-offs among key principles of trustworthy ML, including fairness, privacy, …

• L2GTX: From Local to Global Time Series Explanations

  Ephrem Tibebe Mekonnen, Luca Longo, Lucas Rizzo, Pierpaolo Dondio · 16 mars 2026

  Deep learning models achieve high accuracy in time series classification, yet understanding their class-level decision behaviour remains challenging. Explanations for time series must respect temporal dependencies and identify patterns that recur across instances. Existing approaches face three limi…

• Decomposing Observational Multiplicity in Decision Trees: Leaf and Structural Regret

  Mustafa Cavus · 13 mars 2026

  Many machine learning tasks admit multiple models that perform almost equally well, a phenomenon known as predictive multiplicity. A fundamental source of this multiplicity is observational multiplicity, which arises from the stochastic nature of label collection: observed training labels represent …

• Mitigating the Multiplicity Burden: The Role of Calibration in Reducing Predictive Multiplicity of Classifiers

  Mustafa Cavus · 13 mars 2026

  As machine learning models are increasingly deployed in high-stakes environments, ensuring both probabilistic reliability and prediction stability has become critical. This paper examines the interplay between classification calibration and predictive multiplicity - the phenomenon in which multiple …

• Attention Gathers, MLPs Compose: A Causal Analysis of an Action-Outcome Circuit in VideoViT

  Sai V R Chereddy · 13 mars 2026

  The paper explores how video models trained for classification tasks represent nuanced, hidden semantic information that may not affect the final outcome, a key challenge for Trustworthy AI models. Through Explainable and Interpretable AI methods, specifically mechanistic interpretability techniques…

• Evaluating Explainable AI Attribution Methods in Neural Machine Translation via Attention-Guided Knowledge Distillation

  Aria Nourbakhsh, Salima Lamsiyah, Adelaide Danilov, Christoph Schommer · 13 mars 2026

  The study of the attribution of input features to the output of neural network models is an active area of research. While numerous Explainable AI (XAI) techniques have been proposed to interpret these models, the systematic and automated evaluation of these methods in sequence-to-sequence (seq2seq)…

• COMPASS: The explainable agentic framework for Sovereignty, Sustainability, Compliance, and Ethics

  Jean-S\'ebastien, Dessureault, Alain-Thierry, Iliho Manzi, Soukaina, Alaoui Ismaili, Khadim, Lo, Mireille, Lalancette, \'Eric, B\'elanger · 13 mars 2026

  The rapid proliferation of large language model (LLM)-based agentic systems raises critical concerns regarding digital sovereignty, environmental sustainability, regulatory compliance, and ethical alignment. Whilst existing frameworks address individual dimensions in isolation, no unified architectu…

• FinRule-Bench: A Benchmark for Joint Reasoning over Financial Tables and Principles

  Arun Vignesh Malarkkan, Manan Roy Choudhury, Guangwei Zhang, Vivek Gupta, Qingyun Wang, Yanjie Fu, Denghui Zhang · 13 mars 2026

  Large language models (LLMs) are increasingly applied to financial analysis, yet their ability to audit structured financial statements under explicit accounting principles remains poorly explored. Existing benchmarks primarily evaluate question answering, numerical reasoning, or anomaly detection o…

• Learning Tree-Based Models with Gradient Descent

  Sascha Marton · 13 mars 2026

  Tree-based models are widely recognized for their interpretability and have proven effective in various application domains, particularly in high-stakes domains. However, learning decision trees (DTs) poses a significant challenge due to their combinatorial complexity and discrete, non-differentiabl…

• Examining Reasoning LLMs-as-Judges in Non-Verifiable LLM Post-Training

  Yixin Liu, Yue Yu, DiJia Su, Sid Wang, Xuewei Wang, Song Jiang, Bo Liu, Arman Cohan, Yuandong Tian, Zhengxing Chen · 13 mars 2026

  Reasoning LLMs-as-Judges, which can benefit from inference-time scaling, provide a promising path for extending the success of reasoning models to non-verifiable domains where the output correctness/quality cannot be directly checked. However, while reasoning judges have shown better performance on …

• Why Does It Look There? Structured Explanations for Image Classification

  Jiarui Li, Zixiang Yin, Samuel J Landry, Zhengming Ding, Ramgopal R. Mettu · 12 mars 2026

  Deep learning models achieve remarkable predictive performance, yet their black-box nature limits transparency and trustworthiness. Although numerous explainable artificial intelligence (XAI) methods have been proposed, they primarily provide saliency maps or concepts (i.e., unstructured interpretab…

• Code-Space Response Oracles: Generating Interpretable Multi-Agent Policies with Large Language Models

  Daniel Hennes, Zun Li, John Schultz, Marc Lanctot · 12 mars 2026

  Recent advances in multi-agent reinforcement learning, particularly Policy-Space Response Oracles (PSRO), have enabled the computation of approximate game-theoretic equilibria in increasingly complex domains. However, these methods rely on deep reinforcement learning oracles that produce `black-box'…

• Digging Deeper: Learning Multi-Level Concept Hierarchies

  Oscar Hill, Mateo Espinosa Zarlenga, Mateja Jamnik · 12 mars 2026

  Although concept-based models promise interpretability by explaining predictions with human-understandable concepts, they typically rely on exhaustive annotations and treat concepts as flat and independent. To circumvent this, recent work has introduced Hierarchical Concept Embedding Models (HiCEMs)…

• Causally Grounded Mechanistic Interpretability for LLMs with Faithful Natural-Language Explanations

  Ajay Pravin Mahale · 12 mars 2026

  Mechanistic interpretability identifies internal circuits responsible for model behaviors, yet translating these findings into human-understandable explanations remains an open problem. We present a pipeline that bridges circuit-level analysis and natural language explanations by (i) identifying cau…

• Rethinking the Harmonic Loss via Non-Euclidean Distance Layers

  Maxwell Miller-Golub, Kamil Faber, Marcin Pietron, Panpan Zheng, Pasquale Minervini, Roberto Corizzo · 12 mars 2026

  Cross-entropy loss has long been the standard choice for training deep neural networks, yet it suffers from interpretability limitations, unbounded weight growth, and inefficiencies that can contribute to costly training dynamics. The harmonic loss is a distance-based alternative grounded in Euclide…

• Personalizing explanations of AI-driven hints to users' characteristics: an empirical evaluation

  Vedant Bahel, Harshinee Sriram, Cristina Conati · 12 mars 2026

  The paper extends an existing Intelligent Tutoring System (ITS) that supports students' learning via AI-driven personalized hints and can generate explanations to justify why/how the hints were generated. In this work, we investigate personalizing these hint explanations to students with low levels …