Attribution-Guided Multimodal Deepfake Detection via Cross-Modal Forensic Fingerprints

• URL: http://arxiv.org/abs/2604.26453v1
• Date: Wed, 29 Apr 2026 09:11:13 GMT
• Title: Attribution-Guided Multimodal Deepfake Detection via Cross-Modal Forensic Fingerprints
• Abstract summary: We argue that a detector incapable of identifying how a video was forged is likely learning the wrong signal.<n>Unlike binary detection, attribution-guided learning imposes a stronger geometric constraint on the shared embedding space.<n>We propose the Attribution-Guided Multimodal Deepfake Detection framework, which jointly learns to detect and attribute manipulation.
• Score: 4.598350938853635
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Audio-visual deepfakes have reached a level of realism that makes perceptual detection unreliable, threatening media integrity and biometric security. While multimodal detection has shown promise, most approaches are binary classification tasks that often latch onto dataset-specific artifacts rather than genuine generative traces. We argue that a detector incapable of identifying how a video was forged is likely learning the wrong signal. Unlike binary detection, attribution-guided learning imposes a stronger geometric constraint on the shared embedding space, forcing the model to encode generator-specific forensic content rather than shortcuts. We propose the Attribution-Guided Multimodal Deepfake Detection (AMDD) framework, which jointly learns to detect and attribute manipulation. AMDD treats generator attribution as a structured regularization that constrains representation geometry toward forensically meaningful features. We introduce a Cross-Modal Forensic Fingerprint Consistency (CMFFC) loss to enforce alignment between generator-induced artifacts in visual and audio streams. This exploits the fact that coherent manipulation leaves correlated traces across modalities, grounded in the physical coupling between speech and facial articulation that synthetic pipelines routinely disrupt. Architecturally, we pair a ResNet50 with temporal attention for visual encoding against a pretrained ResNet18 for mel spectrograms, closing the encoder capacity gap found in prior models. On FakeAVCeleb, AMDD achieves 99.7% balanced accuracy and 99.8% AUC with 95.9% attribution accuracy. Cross-dataset evaluation on DeepfakeTIMIT, DFDM, and LAV-DF confirms that real video detection generalizes robustly, while fake detection on unseen generators remains an open challenge that we analyze in depth.

Related papers

• Toward Accountable AI-Generated Content on Social Platforms: Steganographic Attribution and Multimodal Harm Detection [4.6362646650612085]
  generative AI has introduced new challenges in content moderation and digital forensics.<n> benign AI-generated images can be paired with harmful or misleading text, creating difficult-to-detect misuse.<n>We introduce a steganography enabled attribution framework that embeds cryptographically signed identifiers into images at creation time.
  arXiv  Detail & Related papers  (2026-04-12T04:57:25Z)
• SAVe: Self-Supervised Audio-visual Deepfake Detection Exploiting Visual Artifacts and Audio-visual Misalignment [74.05278327933006]
  We propose SAVe, a self-supervised audio-visual deepfake detection framework that learns entirely on authentic videos.<n>SAVE generates on-the-fly, identity-preserving, region-aware self-blended pseudo-manipulations to emulate tampering artifacts.<n>To capture cross-modal evidence, SAVe also models lip-speech synchronization.
  arXiv  Detail & Related papers  (2026-03-26T08:01:35Z)
• ForensicZip: More Tokens are Better but Not Necessary in Forensic Vision-Language Models [108.5042835056188]
  We introduce ForensicZip, a training-free framework that reformulates token compression from a forgery-driven perspective.<n>ForensicZip achieves $2.97times$ speedup and over 90% FLOPs reduction while maintaining state-of-the-art detection performance.
  arXiv  Detail & Related papers  (2026-03-12T17:30:49Z)
• When Detectors Forget Forensics: Blocking Semantic Shortcuts for Generalizable AI-Generated Image Detection [48.75021863792367]
  Vision Foundation Models (VFMs, emphe.g., CLIP) often struggle to generalize to images created using unseen generation pipelines.<n>We propose textbfGeometric Semantic Decoupling (GSD), a parameter-free module that explicitly removes semantic components from learned representations.<n>GSD estimates semantic directions from batch-wise statistics and projects them out via a geometric constraint, forcing the artifact detector to rely on semantic-invariant forensic evidence.
  arXiv  Detail & Related papers  (2026-03-10T06:16:35Z)
• Multi-modal Deepfake Detection and Localization with FPN-Transformer [21.022230340898556]
  We introduce a multi-modal deepfake detection and localization framework based on a Feature Pyramid-Transformer (FPN-Transformer)<n>A multi-scale feature pyramid is constructed through R-TLM blocks with localized attention mechanisms, enabling joint analysis of cross-context temporal dependencies.<n>We evaluate our approach on the test set of the IJCAI'25 DDL-AV benchmark, showing a good performance with a final score of 0.7535.
  arXiv  Detail & Related papers  (2025-11-11T09:33:39Z)
• DiMoDif: Discourse Modality-information Differentiation for Audio-visual Deepfake Detection and Localization [13.840950434728533]
  DiMoDif is an audio-visual deepfake detection framework.<n>It exploits the inter-modality differences in machine perception of speech.<n>It temporally localizes the deepfake forgery.
  arXiv  Detail & Related papers  (2024-11-15T13:47:33Z)
• CapST: Leveraging Capsule Networks and Temporal Attention for Accurate Model Attribution in Deep-fake Videos [9.209808258321559]
  Attributing a deep-fake to its specific generation model or encoder is vital for forensic analysis, enabling source and tailored countermeasures.<n>We investigate the model attribution problem for deep-fake videos using two datasets: Deepfakes from Different Models (DFDM) and GANGen-Detection.<n>We introduce a novel Capsule-Spatial-Cap (CapST) model that integrates a truncated VGG19 network for feature extraction, capsule networks for temporal extraction.
  arXiv  Detail & Related papers  (2023-11-07T08:05:09Z)
• MMNet: Multi-Collaboration and Multi-Supervision Network for Sequential Deepfake Detection [81.59191603867586]
  Sequential deepfake detection aims to identify forged facial regions with the correct sequence for recovery. The recovery of forged images requires knowledge of the manipulation model to implement inverse transformations. We propose Multi-Collaboration and Multi-Supervision Network (MMNet) that handles various spatial scales and sequential permutations in forged face images.
  arXiv  Detail & Related papers  (2023-07-06T02:32:08Z)
• SeeABLE: Soft Discrepancies and Bounded Contrastive Learning for Exposing Deepfakes [7.553507857251396]
  We propose a novel deepfake detector, called SeeABLE, that formalizes the detection problem as a (one-class) out-of-distribution detection task. SeeABLE pushes perturbed faces towards predefined prototypes using a novel regression-based bounded contrastive loss. We show that our model convincingly outperforms competing state-of-the-art detectors, while exhibiting highly encouraging generalization capabilities.
  arXiv  Detail & Related papers  (2022-11-21T09:38:30Z)
• Multimodal Graph Learning for Deepfake Detection [10.077496841634135]
  Existing deepfake detectors face several challenges in achieving robustness and generalization. We propose a novel framework, namely Multimodal Graph Learning (MGL), that leverages information from multiple modalities. Our proposed method aims to effectively identify and utilize distinguishing features for deepfake detection.
  arXiv  Detail & Related papers  (2022-09-12T17:17:49Z)
• Voice-Face Homogeneity Tells Deepfake [56.334968246631725]
  Existing detection approaches contribute to exploring the specific artifacts in deepfake videos. We propose to perform the deepfake detection from an unexplored voice-face matching view. Our model obtains significantly improved performance as compared to other state-of-the-art competitors.
  arXiv  Detail & Related papers  (2022-03-04T09:08:50Z)
• MD-CSDNetwork: Multi-Domain Cross Stitched Network for Deepfake Detection [80.83725644958633]
  Current deepfake generation methods leave discriminative artifacts in the frequency spectrum of fake images and videos. We present a novel approach, termed as MD-CSDNetwork, for combining the features in the spatial and frequency domains to mine a shared discriminative representation.
  arXiv  Detail & Related papers  (2021-09-15T14:11:53Z)
• Beyond the Spectrum: Detecting Deepfakes via Re-Synthesis [69.09526348527203]
  Deep generative models have led to highly realistic media, known as deepfakes, that are commonly indistinguishable from real to human eyes. We propose a novel fake detection that is designed to re-synthesize testing images and extract visual cues for detection. We demonstrate the improved effectiveness, cross-GAN generalization, and robustness against perturbations of our approach in a variety of detection scenarios.
  arXiv  Detail & Related papers  (2021-05-29T21:22:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.