UniShield: Unified Face Attack Detection via KG-Informed Multimodal Reasoning
Published in arXiv, 2026
UniShield proposes a unified face attack detection framework that uses a face attack knowledge graph and multimodal reasoning to jointly recognize physical spoofing and digital forgery with stronger evidence-grounded explanations.
Recommended citation: Li, H., Shi, Y., Wang, H., Gao, Y., Ma, H., Feng, J., & Yu, Z. (2026). UniShield: Unified Face Attack Detection via KG-Informed Multimodal Reasoning. arXiv, 2605.08709. https://arxiv.org/abs/2605.08709