Explainable Artificial Intelligence (XAI)

Derniers papiers

• Quantification of Credal Uncertainty: A Distance-Based Approach

  Xabier Gonzalez-Garcia, Siu Lun Chau, Julian Rodemann, Michele Caprio, Krikamol Muandet, Humberto Bustince, S\'ebastien Destercke, Eyke H\"ullermeier, Yusuf Sale · 31 mars 2026

  Credal sets, i.e., closed convex sets of probability measures, provide a natural framework to represent aleatoric and epistemic uncertainty in machine learning. Yet how to quantify these two types of uncertainty for a given credal set, particularly in multiclass classification, remains underexplored…

• Position: Explainable AI is Causality in Disguise

  Amir-Hossein Karimi · 31 mars 2026

  The demand for Explainable AI (XAI) has triggered an explosion of methods, producing a landscape so fragmented that we now rely on surveys of surveys. Yet, fundamental challenges persist: conflicting metrics, failed sanity checks, and unresolved debates over robustness and fairness. The only consens…

• Principal Prototype Analysis on Manifold for Interpretable Reinforcement Learning

  Bodla Krishna Vamshi, Haizhao Yang · 31 mars 2026

  Recent years have witnessed the widespread adoption of reinforcement learning (RL), from solving real-time games to fine-tuning large language models using human preference data significantly improving alignment with user expectations. However, as model complexity grows exponentially, the interpreta…

• AirQA: A Comprehensive QA Dataset for AI Research with Instance-Level Evaluation

  Tiancheng Huang, Ruisheng Cao, Yuxin Zhang, Zhangyi Kang, Zijian Wang, Chenrun Wang, Yijie Luo, Hang Zheng, Lirong Qian, Lu Chen, Kai Yu · 31 mars 2026

  The growing volume of academic papers has made it increasingly difficult for researchers to efficiently extract key information. While large language models (LLMs) based agents are capable of automating question answering (QA) workflows for scientific papers, there still lacks a comprehensive and re…

• Designing User-Centric Metrics for Evaluation of Counterfactual Explanations

  Firdaus Ahmed Choudhury, Ethan Leicht, Jude Ethan Bislig, Hangzhi Guo, Amulya Yadav · 31 mars 2026

  Counterfactual Explanations (CFEs) have grown in popularity as a means of offering actionable guidance by identifying the minimum changes in feature values required to flip an ML model's prediction to something more desirable. Unfortunately, most prior research on CFEs relies on artificial evaluatio…

• Low-Rank Adaptation Reduces Catastrophic Forgetting in Sequential Transformer Encoder Fine-Tuning: Controlled Empirical Evidence and Frozen-Backbone Representation Probes

  Ashish Pandey · 31 mars 2026

  Sequential fine-tuning of pretrained language encoders often overwrites previously acquired capabilities, but the forgetting behavior of parameter-efficient updates remains under-characterized. We present a controlled empirical study of Low-Rank Adaptation (LoRA) in sequential transformer encoder fi…

• Conditional Factuality Controlled LLMs with Generalization Certificates via Conformal Sampling

  Kai Ye, Qingtao Pan, Shuo Li · 31 mars 2026

  Large language models (LLMs) need reliable test-time control of hallucinations. Existing conformal methods for LLMs typically provide only \emph{marginal} guarantees and rely on a single global threshold, which can under-cover hard prompts, over-cover easy ones, and produce oversized prediction sets…

• What-If Explanations Over Time: Counterfactuals for Time Series Classification

  Udo Schlegel, Thomas Seidl · 31 mars 2026

  Counterfactual explanations emerge as a powerful approach in explainable AI, providing what-if scenarios that reveal how minimal changes to an input time series can alter the model's prediction. This work presents a survey of recent algorithms for counterfactual explanations for time series classifi…

• Omni-Modal Dissonance Benchmark: Systematically Breaking Modality Consensus to Probe Robustness and Calibrated Abstention

  Zabir Al Nazi, Shubhashis Roy Dipta, Md Rizwan Parvez · 31 mars 2026

  Existing omni-modal benchmarks attempt to measure modality-specific contributions, but their measurements are confounded: naturally co-occurring modalities carry correlated yet unequal information, making it unclear whether results reflect true modality reliance or information asymmetry. We introduc…

• Strategic Candidacy in Generative AI Arenas

  Chris Hays, Rachel Li, Bailey Flanigan, Manish Raghavan · 31 mars 2026

  AI arenas, which rank generative models from pairwise preferences of users, are a popular method for measuring the relative performance of models in the course of their organic use. Because rankings are computed from noisy preferences, there is a concern that model producers can exploit this randomn…

• Beyond the Answer: Decoding the Behavior of LLMs as Scientific Reasoners

  Rohan Pandey, Eric Ye, Michael Li · 31 mars 2026

  As Large Language Models (LLMs) achieve increasingly sophisticated performance on complex reasoning tasks, current architectures serve as critical proxies for the internal heuristics of frontier models. Characterizing emergent reasoning is vital for long-term interpretability and safety. Furthermore…

• CLMN: Concept based Language Models via Neural Symbolic Reasoning

  Yibo Yang · 31 mars 2026

  Deep learning has advanced NLP, but interpretability remains limited, especially in healthcare and finance. Concept bottleneck models tie predictions to human concepts in vision, but NLP versions either use binary activations that harm text representations or latent concepts that weaken semantics, a…

• A Regression Framework for Understanding Prompt Component Impact on LLM Performance

  Andrew Lauziere, Jonathan Daugherty, Taisa Kushner · 31 mars 2026

  As large language models (LLMs) continue to improve and see further integration into software systems, so does the need to understand the conditions in which they will perform. We contribute a statistical framework for understanding the impact of specific prompt features on LLM performance. The appr…

• MonitorBench: A Comprehensive Benchmark for Chain-of-Thought Monitorability in Large Language Models

  Han Wang, Yifan Sun, Brian Ko, Mann Talati, Jiawen Gong, Zimeng Li, Naicheng Yu, Xucheng Yu, Wei Shen, Vedant Jolly, Huan Zhang · 31 mars 2026

  Large language models (LLMs) can generate chains of thought (CoTs) that are not always causally responsible for their final outputs. When such a mismatch occurs, the CoT no longer faithfully reflects the decision-critical factors driving the model's behavior, leading to the reduced CoT monitorabilit…

• Concerning Uncertainty -- A Systematic Survey of Uncertainty-Aware XAI

  Helena L\"ofstr\"om, Tuwe L\"ofstr\"om, Anders Hjort, Fatima Rabia Yapicioglu · 31 mars 2026

  This paper surveys uncertainty-aware explainable artificial intelligence (UAXAI), examining how uncertainty is incorporated into explanatory pipelines and how such methods are evaluated. Across the literature, three recurring approaches to uncertainty quantification emerge (Bayesian, Monte Carlo, an…

• Stable Reasoning, Unstable Responses: Mitigating LLM Deception via Stability Asymmetry

  Guoxi Zhang, Jiawei Chen, Tianzhuo Yang, Lang Qin, Juntao Dai, Yaodong Yang, Jingwei Yi · 31 mars 2026

  As Large Language Models (LLMs) expand in capability and application scope, their trustworthiness becomes critical. A vital risk is intrinsic deception, wherein models strategically mislead users to achieve their own objectives. Existing alignment approaches based on chain-of-thought (CoT) monitorin…

• Coherent Without Grounding, Grounded Without Success: Observability and Epistemic Failure

  Camilo Chac\'on Sartori · 31 mars 2026

  When an agent can articulate why something works, we typically take this as evidence of genuine understanding. This presupposes that effective action and correct explanation covary, and that coherent explanation reliably signals both. I argue that this assumption fails for contemporary Large Languag…

• ORACAL: A Robust and Explainable Multimodal Framework for Smart Contract Vulnerability Detection with Causal Graph Enrichment

  Tran Duong Minh Dai, Triet Huynh Minh Le, M. Ali Babar, Van-Hau Pham, Phan The Duy · 31 mars 2026

  Although Graph Neural Networks (GNNs) have shown promise for smart contract vulnerability detection, they still face significant limitations. Homogeneous graph models fail to capture the interplay between control flow and data dependencies, while heterogeneous graph approaches often lack deep semant…

• Evaluating Interactive 2D Visualization as a Sample Selection Strategy for Biomedical Time-Series Data Annotation

  Einari Vaaras, Manu Airaksinen, Okko R\"as\"anen · 30 mars 2026

  Reliable machine-learning models in biomedical settings depend on accurate labels, yet annotating biomedical time-series data remains challenging. Algorithmic sample selection may support annotation, but evidence from studies involving real human annotators is scarce. Consequently, we compare three …

• "What don't you understand?" Language games and black box algorithms

  Remy Demichelis · 30 mars 2026

  The aim of this article is to understand the problem of "black box" algorithms, an issue inherent to the nascent field of Explainable Artificial Intelligence (XAI). While it is relatively easy to understand something someone explained to us, it becomes more complicated when no one can fully grasp th…

• SALMUBench: A Benchmark for Sensitive Association-Level Multimodal Unlearning

  Cai Selvas-Sala, Lei Kang, Lluis Gomez · 30 mars 2026

  As multimodal models like CLIP become integral to downstream systems, the need to remove sensitive information is critical. However, machine unlearning for contrastively-trained encoders remains underexplored, and existing evaluations fail to diagnose fine-grained, association-level forgetting. We i…

• Do Language Models Follow Occam's Razor? An Evaluation of Parsimony in Inductive and Abductive Reasoning

  Yunxin Sun, Abulhair Saparov · 27 mars 2026

  Non-deductive reasoning, encompassing inductive and abductive reasoning, is essential in addressing complex real-world questions. One key feature of inductive and abductive reasoning is that there are many valid hypotheses; the simplest ones (those that adhere to Occam's Razor) are often most useful…

• Do LLMs Know What They Know? Measuring Metacognitive Efficiency with Signal Detection Theory

  Jon-Paul Cacioli · 27 mars 2026

  Standard evaluation of LLM confidence relies on calibration metrics (ECE, Brier score) that conflate two distinct capacities: how much a model knows (Type-1 sensitivity) and how well it knows what it knows (Type-2 metacognitive sensitivity). We introduce an evaluation framework based on Type-2 Signa…

• A Practical Guide Towards Interpreting Time-Series Deep Clinical Predictive Models: A Reproducibility Study

  Yongda Fan, John Wu, Andrea Fitzpatrick, Naveen Baskaran, Jimeng Sun, Adam Cross · 27 mars 2026

  Clinical decisions are high-stakes and require explicit justification, making model interpretability essential for auditing deep clinical models prior to deployment. As the ecosystem of model architectures and explainability methods expands, critical questions remain: Do architectural features like …

• More Than "Means to an End": Supporting Reasoning with Transparently Designed AI Data Science Processes

  Venkatesh Sivaraman, Patrick Vossler, Adam Perer, Julian Hong, Jean Feng · 27 mars 2026

  Generative artificial intelligence (AI) tools can now help people perform complex data science tasks regardless of their expertise. While these tools have great potential to help more people work with data, their end-to-end approach does not support users in evaluating alternative approaches and ref…