explainable ai and reasoning

This work addresses the theoretical shortcomings of prevailing post-hoc feature attribution methods—such as SHAP—which often lack formal guarantees and can mislead human decision-making in high-stakes settings. To overcome this limitation, the paper introduces a novel paradigm grounded in symbolic explainable artificial intelligence (XAI), leveraging formal, verifiable symbolic reasoning to reconstruct feature importance assignments. By integrating formal verification with feature attribution analysis, the proposed framework establishes a theoretically sound and certifiable approach to interpretability. This integration significantly enhances the reliability and trustworthiness of XAI in safety-critical applications, offering a rigorous and verifiable pathway for generating explanations in high-risk machine learning systems.

This paper identifies the “explanation inversion” problem in posteriori explanation methods (e.g., LIME, SHAP): such methods often rationalize model predictions *backward* from outputs rather than faithfully reconstructing the true input-to-output decision pathway—especially under spurious correlations. To address this, the authors propose Inversion Quantification (IQ), the first formal, quantifiable framework for measuring explanation inversion. They further introduce Reproduce-by-Poking (RBP), a model-agnostic enhancement strategy that applies forward perturbations to inputs and theoretically guarantees inversion mitigation. Extensive evaluation across tabular, image, and text modalities on synthetic benchmarks demonstrates that mainstream explanation methods consistently exhibit inversion; RBP reduces inversion by an average of 1.8% while offering both empirical efficacy and theoretical provability.

This study addresses the lack of a domain-agnostic, human-centered explainable artificial intelligence (XAI) framework by investigating user preferences across healthcare, retail, and energy domains. Through expert interviews and multi-stakeholder structured surveys, it empirically identifies “interpretability over accuracy” as a cross-domain preference and establishes feature importance and counterfactual explanations as the two foundational pillars of a universal XAI framework. Method: The approach integrates qualitative transcription analysis, questionnaire-driven requirement modeling, and genetic programming (GP) to construct inherently interpretable models. Contribution/Results: We propose the first empirically validated, unified XAI framework spanning multiple domains; release an open-source, standardized XAI questionnaire toolkit; demonstrate its feasibility across three core machine learning tasks—prediction, diagnosis, and prescription—and advance the XAI paradigm from technology-centric design toward human consensus–driven development.

This study addresses the troubling inconsistency in feature attribution methods such as SHAP, which can yield substantially divergent explanations even for identical inputs and models, thereby undermining trustworthiness and auditability in high-stakes applications. The work formally defines and quantifies the phenomenon of “explanation multiplicity,” distinguishing its origins in model training or selection from inherent randomness in the explanation procedure itself. To assess stability, the authors introduce a dual-perspective metric incorporating both feature magnitude and ranking, establish a randomized null model as an interpretable baseline, and develop a comprehensive empirical evaluation framework spanning diverse datasets and model classes. Experiments demonstrate that explanation multiplicity is pervasive; relying solely on SHAP value magnitudes can lead to misleading conclusions, necessitating rank-sensitive metrics and principled baselines for reliable interpretability assessment.

This study addresses the question of what user-relevant content should be included in local, post-hoc explanations for industrial AI systems and how such content should be organized. Through a hybrid inductive–deductive qualitative content analysis, the authors integrate user research data from six domains with explainable AI theory to develop the first user-oriented model comprising fourteen categories of local explanation content. The model encompasses rule-based and causal dimensions alongside two cognitive dimensions. It demonstrates high content validity and clear conceptual boundaries, as evidenced by expert review, an item-level content validity index (I-CVI ≥ 0.82), and strong intercoder reliability (Krippendorff’s α = Cohen’s κ = 0.920). This framework provides both theoretical grounding and practical guidance for the elicitation, standardization, and evaluation of explanation content in industrial AI applications.

This study addresses the persistent challenge that explainable AI (XAI) often fails to effectively support human decision-making due to poor user comprehension. To bridge this gap, the work integrates cognitive modeling with user studies to formally represent—within a computationally tractable framework—the reasoning strategies humans employ when interacting with different XAI methods in structured data tasks. Through formative and summative user experiments, feature attribution analyses, and behavioral alignment evaluations, the resulting cognitive model demonstrates significantly greater accuracy than conventional machine learning surrogates in capturing human forward-simulation decision behavior. Beyond elucidating which XAI mechanisms genuinely aid human judgment, the model offers an empirically grounded foundation for designing more effective XAI systems and serves as a high-fidelity, efficient proxy for human-subject experimentation in XAI research.

This work addresses the challenge that logic-based explanations generated by symbolic AI often impose excessive cognitive load on human users due to redundant details, thereby hindering comprehension. To mitigate this issue, the authors propose a simplification approach grounded in formal abstraction, systematically distinguishing and empirically evaluating two strategies—deletion and clustering—in the context of Answer Set Programming (ASP) explanations. Cognitive experiments demonstrate that clustering significantly enhances users’ understanding, while deletion effectively reduces cognitive effort. This study provides the first empirical evidence of the differential impacts of distinct abstraction strategies on human-centered symbolic explanations, offering both theoretical insights and methodological guidance for designing explainable AI systems that balance interpretability with cognitive efficiency.

This study addresses the challenge that existing algorithmic explanations are often poorly understood and misapplied by non-expert users due to semantic ambiguity and insufficient contextual information, leading to a disconnect between explanations and actual decision-making. To bridge this gap, the authors propose an “Explanation Card” framework that augments widely used interpretability methods—such as SHAP and counterfactual explanations—with structured metadata specifying their applicability boundaries, robustness properties, and user-oriented interpretation guidance. By shifting explanatory responsibility from end users to explanation providers, this approach enhances the practical utility and regulatory compliance of model explanations, aligning with the transparency requirements of the EU AI Act. Empirical evaluations demonstrate that Explanation Cards significantly improve users’ comprehension accuracy of complex model explanations and effectively flag scenarios where explanations are unreliable, thereby facilitating trustworthy real-world deployment of algorithmic systems.

This study addresses the challenge that local interpretability methods often produce seemingly plausible yet unfaithful explanations for complex tabular data. To rigorously evaluate explanation fidelity, robustness, and complexity, the authors construct a comprehensive benchmarking framework encompassing multiple models and datasets, incorporating for the first time a prediction-consistency grouping strategy. They systematically assess prominent methods—including LIME, Kernel SHAP, and feature ablation—across 32 tabular datasets. The findings reveal that explanation quality shows no significant correlation with model accuracy; instead, it is predominantly influenced by data complexity and feature distribution, particularly on samples where models consistently err. This work provides a novel perspective and empirical foundation for trustworthy evaluation in explainable AI.