mlsanityData to 7 Oct 2026

UniShield: Unified Face Attack Detection via KG-Informed Multimodal Reasoning

arXiv 2605.08709 ↗First posted 9 May 2026cs.CVcited by 0

On the map: LLM reasoning → Multimodal reasoning

Abstract (arXiv)

Unified face attack detection (UAD) requires recognizing physical spoofing and digital forgery within a shared decision space, yet existing discriminative or prompt-based methods largely rely on appearance correlations and provide limited evidence-grounded reasoning. We propose UniShield, a knowledge-grounded multimodal reasoning framework for unified face attack defense. UniShield constructs a Face Attack Knowledge Graph (FAKG) that links attack categories to diagnostic visual cues and attack-conditioned relations, and uses it to synthesize 52,025 FAKG-QA examples for Attack-Graph Instruction Tuning (AGIT). To improve rationale consistency, we further introduce Graph-Consistent Reasoning Optimization (GCRO), a GRPO-based objective with a KG-consistency reward that encourages generated rationales to match graph-supported cues while penalizing incompatible claims. Experiments on our multimodal UAD benchmark show that UniShield achieves strong performance across binary, coarse-grained, and fine-grained protocols, with consistently high ACC and low HTER. These results suggest that structured attack knowledge can improve both detection accuracy and reasoning reliability over discriminative baselines and general-purpose MLLMs. Our code will be released at https://anonymous.4open.science/r/Unishield-A6A3/.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 22 of its 33 references that are in the map. First posted on arXiv 9 May 2026.

Builds on

22 of 33 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning1 Jul
  2. 2025Harnessing Chain-of-Thought Reasoning in Multimodal Large Language Models for Face Anti-Spoofing2 Jun
  3. 2024SHIELD : An Evaluation Benchmark for Face Spoofing and Forgery Detection with Multimodal Large Language Models6 Feb

Other references · 19

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,069
  3. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,689
  4. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,179
  5. 2025OpenAI GPT-5 System Cardcited by 910
  6. 2019Face X-ray for More General Face Forgery Detectioncited by 469
  7. 2018Capsule-Forensics: Using Capsule Networks to Detect Forged Images and Videoscited by 180
  8. 2020Searching Central Difference Convolutional Networks for Face Anti-Spoofingcited by 173
  9. 2014Learn Convolutional Neural Network for Face Anti-Spoofingcited by 80
  10. 2024CFPL-FAS: Class Free Prompt Learning for Generalizable Face Anti-spoofingcited by 39
  11. 2022Benchmarking Joint Face Spoofing and Forgery Detection with Visual and Physiological Cuescited by 24
  12. 2025Unlocking the Capabilities of Large Vision-Language Models for Generalizable and Explainable Deepfake Detectioncited by 20
7 more
  1. 2024Unified Physical-Digital Face Attack Detectioncited by 17
  2. 2025VLForgery Face Triad: Detection, Localization and Attribution via Multimodal Large Language Modelscited by 12
  3. 2024La-SoftMoE CLIP for Unified Physical-Digital Face Attack Detectioncited by 8
  4. 2025GraphGen: Enhancing Supervised Fine-Tuning for LLMs with Knowledge-Driven Synthetic Data Generationcited by 7
  5. 2025FaceShield: Explainable Face Anti-Spoofing with Multimodal Large Language Modelscited by 6
  6. 2025Benchmarking Unified Face Attack Detection via Hierarchical Prompt Tuningcited by 5
  7. 2025Knowledge-Guided Prompt Learning for Deepfake Facial Image Detectioncited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026MS-MFAD : Multimodal large language models for Face Anti-spoofing Detection
  2. 2026FAS-R1: A Unified Multi-Task MLLM for Reasoning Face Anti-Spoofing
  3. 2026From Intuition to Investigation: A Tool-Augmented Reasoning MLLM Framework for Generalizable Face Anti-Spoofing
  4. 2025PA-FAS: Towards Interpretable and Generalizable Multimodal Face Anti-Spoofing via Path-Augmented Reinforcement Learning
  5. 2025Exploring Task-Solving Paradigm for Generalized Cross-Domain Face Anti-Spoofing via Reinforcement Fine-Tuning
  6. 2025InstructFLIP: Exploring Unified Vision-Language Model for Face Anti-spoofing
  7. 2025Poisoned-MRAG: Knowledge Poisoning Attacks to Multimodal Retrieval Augmented Generation
  8. 2025SUEDE:Shared Unified Experts for Physical-Digital Face Attack Detection Enhancement