mlsanityData to 7 Oct 2026

Leveraging large multimodal models for audio-video deepfake detection: a pilot study

arXiv 2602.23393 ↗First posted 25 Feb 2026cs.SD cs.CVcited by 0

On the map: Multimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness

Abstract (arXiv)

Audio-visual deepfake detection (AVD) is increasingly important as modern generators can fabricate convincing speech and video. Most current multimodal detectors are small, task-specific models: they work well on curated tests but scale poorly and generalize weakly across domains. We introduce AV-LMMDetect, a supervised fine-tuned (SFT) large multimodal model that casts AVD as a prompted yes/no classification - "Is this video real or fake?". Built on Qwen 2.5 Omni, it jointly analyzes audio and visual streams for deepfake detection and is trained in two stages: lightweight LoRA alignment followed by audio-visual encoder full fine-tuning. On FakeAVCeleb and Mavos-DD, AV-LMMDetect matches or surpasses prior methods and sets a new state of the art on Mavos-DD datasets.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 15 of its 19 references that are in the map. First posted on arXiv 25 Feb 2026.

Builds on

15 of 19 references.

References · 15

  1. 2016Xception: Deep Learning with Depthwise Separable Convolutionscited by 2,320
  2. 2025Qwen2.5-Omni Technical Reportcited by 704
  3. 2018MesoNet: a Compact Facial Video Forgery Detection Networkcited by 409
  4. 2021Multi-attentional Deepfake Detectioncited by 339
  5. 2020Lips Don't Lie: A Generalisable and Robust Approach to Face Forgery Detectioncited by 286
  6. 2018Capsule-Forensics: Using Capsule Networks to Detect Forged Images and Videoscited by 180
  7. 2021FakeAVCeleb: A Novel Audio-Video Multimodal Deepfake Datasetcited by 179
  8. 2022Self-supervised Learning of Adversarial Example: Towards Good Generalizations for Deepfake Detectioncited by 172
  9. 2020Emotions Don't Lie: An Audio-Visual Deepfake Detection Method Using Affective Cuescited by 111
  10. 2023TALL: Thumbnail Layout for Deepfake Video Detectioncited by 100
  11. 2024AVFF: Audio-Visual Feature Fusion for Video Deepfake Detectioncited by 59
  12. 2022Voice-Face Homogeneity Tells Deepfakecited by 44
3 more
  1. 2024Cross-Modality and Within-Modality Regularization for Audio-Visual DeepFake Detectioncited by 23
  2. 2025ALLM4ADD: Unlocking the Capabilities of Audio Large Language Models for Audio Deepfake Detectioncited by 17
  3. 2025MAVOS-DD: Multilingual Audio-Video Open-Set Deepfake Detection Benchmarkcited by 9

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026X-AVDT: Audio-Visual Cross-Attention for Robust Deepfake Detection
  2. 2026Investigating the Viability of Employing Multi-modal Large Language Models in the Context of Audio Deepfake Detection
  3. 2025HOLA: Enhancing Audio-visual Deepfake Detection via Hierarchical Contextual Aggregations and Efficient Pre-training
  4. 2026ToolDF: Tool-Integrated Reasoning for Mixed-Authenticity Audio Deepfake Detection
  5. 2026DF-MoE: Generalizable Deepfake Detection via Multimodal Sparse Mixture-of-Experts
  6. 2024Evolving from Single-modal to Multi-modal Facial Deepfake Detection: Progress and Challenges
  7. 2026Interpretable All-Type Audio Deepfake Detection with Audio LLMs via Frequency-Time Reinforcement Learning
  8. 2025EDVD-LLaMA: Explainable Deepfake Video Detection via Multimodal Large Language Model Reasoning