VEXA: Evidence-Grounded and Persona-Adaptive Explanations for Scam Risk Sensemaking

TL;DR

VEXA is a framework that generates trustworthy, evidence-based, and persona-adaptive explanations for scam risk detection, enhancing interpretability and user trust in multi-channel online scam scenarios.

Contribution

VEXA introduces a novel approach combining evidence grounding with persona adaptation to improve scam explanation clarity and trustworthiness.

Findings

Grounded explanations improve semantic reliability.

Persona conditioning adds interpretable stylistic variation.

Evidential grounding ensures semantic correctness.

Abstract

Online scams across email, short message services, and social media increasingly challenge everyday risk assessment, particularly as generative AI enables more fluent and context-aware deception. Although transformer-based detectors achieve strong predictive performance, their explanations are often opaque to non-experts or misaligned with model decisions. We propose VEXA, an evidence-grounded and persona-adaptive framework for generating learner-facing scam explanations by integrating GradientSHAP-based attribution with theory-informed vulnerability personas. Evaluation across multi-channel datasets shows that grounding explanations in detector-derived evidence improves semantic reliability without increasing linguistic complexity, while persona conditioning introduces interpretable stylistic variation without disrupting evidential alignment. These results reveal a key design insight: evidential grounding governs semantic correctness, whereas persona-based adaptation operates at the level of presentation under constraints of faithfulness. Together, VEXA demonstrates the feasibility of persona-adaptive, evidence-grounded explanations and provides design guidance for trustworthy, learner-facing security explanations in non-formal contexts.