XAI-Grounded Explanation Generation for Speech Deepfake Detection with Training-Free Multimodal Large Language Models
2026-06-15 • Computation and Language
Computation and LanguageArtificial Intelligence
AI summaryⓘ
The authors address the problem of explaining how speech deepfake detection (SDD) systems make decisions. They note that traditional explanation methods are often low-level and hard for humans to understand, while explanations generated by large language models (LLMs) can be vague and unsupported by evidence. To fix this, the authors propose a new method that combines evidence from explainable AI (XAI) with LLMs to create more specific and grounded explanations without needing additional training. They tested their approach on a dataset called PartialSpoof and found it significantly improved explanation accuracy and reliability, as confirmed by human reviews.
Speech Deepfake DetectionExplainable AIGradient-based AttributionLarge Language ModelsMultimodal ModelsGrounded ExplanationsPartialSpoof DatasetFaithfulness EvaluationHuman EvaluationTraining-free Framework
Authors
Yupei Li, Qiyang Sun, Xiaoliang Wu, Chenxi Wang, Berrak Sisman, Björn W. Schuller
Abstract
Speech deepfake detection (SDD) systems require trustworthy explanations for reliable decision-making. Existing explanation ways mainly fall into two categories. Traditional explainable AI (XAI), such as gradient-based attribution, produces low-level attribution signals tightly coupled with model decisions, and harder to be understood by human than natural language explanations. Meanwhile, large language model (LLM)-based explanation generation often produces generic and ungrounded descriptions due to the lack of heuristic evidence and task-specific supervision, stemming from limited grounded explanation datasets for SDD. We therefore propose a training-free explanation framework that integrates XAI evidence with multimodal LLMs to generate grounded and specific explanations. Using the PartialSpoof dataset, we construct a grounded explanation dataset and show that methods with XAI increase inside accuracy by over 45\%, verified through human evaluation and faithfulness checks.