Unraveling Hidden Representations: A Multi-Modal Layer Analysis for Better Synthetic Content Forensics

• URL: http://arxiv.org/abs/2508.00784v1
• Date: Fri, 01 Aug 2025 17:07:00 GMT
• Title: Unraveling Hidden Representations: A Multi-Modal Layer Analysis for Better Synthetic Content Forensics
• Authors: Tom Or, Omri Azencot,
• Abstract summary: malicious users exploit synthetic media for spreading misinformation and disseminating deepfakes.<n>Need for robust and stable fake detectors is pressing, especially when new generative models appear everyday.<n>We propose the use of large pre-trained multi-modal models for the detection of generative content.
• Score: 4.910937238451485
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Generative models achieve remarkable results in multiple data domains, including images and texts, among other examples. Unfortunately, malicious users exploit synthetic media for spreading misinformation and disseminating deepfakes. Consequently, the need for robust and stable fake detectors is pressing, especially when new generative models appear everyday. While the majority of existing work train classifiers that discriminate between real and fake information, such tools typically generalize only within the same family of generators and data modalities, yielding poor results on other generative classes and data domains. Towards a universal classifier, we propose the use of large pre-trained multi-modal models for the detection of generative content. Effectively, we show that the latent code of these models naturally captures information discriminating real from fake. Building on this observation, we demonstrate that linear classifiers trained on these features can achieve state-of-the-art results across various modalities, while remaining computationally efficient, fast to train, and effective even in few-shot settings. Our work primarily focuses on fake detection in audio and images, achieving performance that surpasses or matches that of strong baseline methods.

Related papers

• Seeing What Matters: Generalizable AI-generated Video Detection with Forensic-Oriented Augmentation [18.402668470092294]
  Synthetic video generation can produce very realistic high-resolution videos that are virtually indistinguishable from real ones.<n>Several video forensic detectors have been recently proposed, but they often exhibit poor generalization.<n>We introduce a novel data augmentation strategy based on the wavelet decomposition and replace specific frequency-related bands to drive the model to exploit more relevant forensic cues.<n>Our method achieves a significant accuracy improvement over state-of-the-art detectors and obtains excellent results even on very recent generative models.
  arXiv  Detail & Related papers  (2025-06-20T07:36:59Z)
• Do DeepFake Attribution Models Generalize? [8.198174674722067]
  Recent advancements in DeepFake generation, along with the proliferation of open-source tools, have significantly lowered the barrier for creating synthetic media.<n>In this work, we leverage five state-of-the-art backbone models and conduct extensive experiments across six DeepFake datasets.<n>Our findings indicate that while binary models demonstrate better generalization abilities, larger models, contrastive methods, and higher data quality can lead to performance improvements in attribution models.
  arXiv  Detail & Related papers  (2025-05-22T13:49:05Z)
• A Self-Learning Multimodal Approach for Fake News Detection [35.98977478616019]
  We introduce a self-learning multimodal model for fake news classification.<n>The model leverages contrastive learning, a robust method for feature extraction that operates without requiring labeled data.<n>Our experimental results on a public dataset demonstrate that the proposed model outperforms several state-of-the-art classification approaches.
  arXiv  Detail & Related papers  (2024-12-08T07:41:44Z)
• Understanding and Improving Training-Free AI-Generated Image Detections with Vision Foundation Models [68.90917438865078]
  Deepfake techniques for facial synthesis and editing pose serious risks for generative models.<n>In this paper, we investigate how detection performance varies across model backbones, types, and datasets.<n>We introduce Contrastive Blur, which enhances performance on facial images, and MINDER, which addresses noise type bias, balancing performance across domains.
  arXiv  Detail & Related papers  (2024-11-28T13:04:45Z)
• Training-Free Deepfake Voice Recognition by Leveraging Large-Scale Pre-Trained Models [52.04189118767758]
  Generalization is a main issue for current audio deepfake detectors. In this paper we study the potential of large-scale pre-trained models for audio deepfake detection.
  arXiv  Detail & Related papers  (2024-05-03T15:27:11Z)
• Data-Independent Operator: A Training-Free Artifact Representation Extractor for Generalizable Deepfake Detection [105.9932053078449]
  In this work, we show that, on the contrary, the small and training-free filter is sufficient to capture more general artifact representations. Due to its unbias towards both the training and test sources, we define it as Data-Independent Operator (DIO) to achieve appealing improvements on unseen sources. Our detector achieves a remarkable improvement of $13.3%$, establishing a new state-of-the-art performance.
  arXiv  Detail & Related papers  (2024-03-11T15:22:28Z)
• Learning Defect Prediction from Unrealistic Data [57.53586547895278]
  Pretrained models of code have become popular choices for code understanding and generation tasks. Such models tend to be large and require commensurate volumes of training data. It has become popular to train models with far larger but less realistic datasets, such as functions with artificially injected bugs. Models trained on such data tend to only perform well on similar data, while underperforming on real world programs.
  arXiv  Detail & Related papers  (2023-11-02T01:51:43Z)
• Convolutional autoencoder-based multimodal one-class classification [80.52334952912808]
  One-class classification refers to approaches of learning using data from a single class only. We propose a deep learning one-class classification method suitable for multimodal data.
  arXiv  Detail & Related papers  (2023-09-25T12:31:18Z)
• Leaving Reality to Imagination: Robust Classification via Generated Datasets [24.411444438920988]
  Recent research on robustness has revealed significant performance gaps between neural image classifiers trained on datasets similar to the test set. We study the question: How do generated datasets influence the natural robustness of image classifiers? We find that Imagenet classifiers trained on real data augmented with generated data achieve higher accuracy and effective robustness than standard training.
  arXiv  Detail & Related papers  (2023-02-05T22:49:33Z)
• Robust Task-Oriented Dialogue Generation with Contrastive Pre-training and Adversarial Filtering [17.7709632238066]
  Data artifacts incentivize machine learning models to learn non-transferable generalizations. We investigate whether popular datasets such as MultiWOZ contain such data artifacts. We propose a contrastive learning based framework to encourage the model to ignore these cues and focus on learning generalisable patterns.
  arXiv  Detail & Related papers  (2022-05-20T03:13:02Z)
• Laplacian Denoising Autoencoder [114.21219514831343]
  We propose to learn data representations with a novel type of denoising autoencoder. The noisy input data is generated by corrupting latent clean data in the gradient domain. Experiments on several visual benchmarks demonstrate that better representations can be learned with the proposed approach.
  arXiv  Detail & Related papers  (2020-03-30T16:52:39Z)

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.