HyperPotter: Spell the Charm of High-Order Interactions in Audio Deepfake Detection

2026-02-05

AI Summary

Existing audio deepfake detection methods often rely on local time-frequency features or pairwise relationships, which struggle to capture discriminative forgery traces embedded in high-order interactions. To address this limitation, this work proposes HyperPotter, a novel framework that introduces hypergraph neural networks to the task for the first time. HyperPotter explicitly models high-order collaborative relationships among multidimensional features by generating learnable hyperedges through clustering and incorporating class-aware prototype initialization. The proposed method achieves an average relative improvement of 22.15% across 11 datasets and outperforms the current state-of-the-art by 13.96% on four cross-domain challenge sets, demonstrating significantly enhanced generalization against diverse forgery attacks and unseen speakers.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Adversarial Attacks & Robustness
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Advances in AIGC technologies have enabled the synthesis of highly realistic audio deepfakes capable of deceiving human auditory perception. Although numerous audio deepfake detection (ADD) methods have been developed, most rely on local temporal/spectral features or pairwise relations, overlooking high-order interactions (HOIs). HOIs capture discriminative patterns that emerge from multiple feature components beyond their individual contributions. We propose HyperPotter, a hypergraph-based framework that explicitly models these synergistic HOIs through clustering-based hyperedges with class-aware prototype initialization. Extensive experiments demonstrate that HyperPotter surpasses its baseline by an average relative gain of 22.15% across 11 datasets and outperforms state-of-the-art methods by 13.96% on 4 challenging cross-domain datasets, demonstrating superior generalization to diverse attacks and speakers.

Problem

• audio deepfake detection
• high-order interactions
• AIGC
• deepfake detection
• feature interactions

Innovation

• high-order interactions
• hypergraph
• audio deepfake detection
• class-aware prototype
• cross-domain generalization