{"slug": "spectra-multi-scale-implicit-decomposition-for-deepfake-detection", "title": "Spectra: Multi-Scale Implicit Decomposition for Deepfake Detection", "summary": "Researchers posted arXiv:2610.31979v1, a paper proposing Spectra, a deepfake detection framework that fits an implicit neural representation (INR) to each image and analyzes its internal activations to expose reconstruction dynamics rather than RGB appearance or frequency-domain artifacts. Spectra extracts multi-level reconstruction heatmaps from image-specific INRs and fuses them bidirectionally with semantic visual features, which the authors report outperforms detectors based solely on RGB appearance or semantic representations across multiple deepfake benchmarks with improved cross-dataset generalization. The authors also report the representation extends to face presentation attack detection, positioning it as a unified framework for facial attack detection.", "body_md": "arXiv:2610.31979v1 Announce Type: new \nAbstract: Deepfakes increasingly undermine trust in digital media, yet existing detectors often struggle to generalize across unseen manipulation techniques, compression levels, and dataset shifts. Most approaches rely on RGB appearance, frequency-domain artifacts, or their combination, learning cues present in the final synthesized image. While effective in-distribution, such cues are frequently generator-specific and can diminish as synthesis models evolve. We argue that a complementary perspective can be obtained through implicit neural representations (INRs), which expose how an image is reconstructed rather than merely how it appears. By fitting an INR to an image and analyzing its internal activations, one can reveal reconstruction dynamics that are difficult to observe in the image domain alone and may capture more intrinsic differences between genuine and synthetic content. To this end, we propose Spectra, a deepfake detection framework that combines semantic visual representations with artifact-aware cues derived from INR reconstruction. Spectra extracts multi-level reconstruction heatmaps from image-specific INRs and jointly models them with semantic features through a bidirectional fusion mechanism. This enables direct interaction between image semantics and reconstruction artifacts, allowing the model to identify forgery cues that are less dependent on specific generation pipelines. Extensive experiments on multiple deepfake benchmarks demonstrate improved robustness and cross-dataset generalization, consistently outperforming detectors based solely on RGB appearance or semantic representations. Furthermore, we show that the proposed representation extends beyond deepfake detection to face presentation attack detection, highlighting its potential as a unified framework for facial attack detection.", "url": "https://wpnews.pro/news/spectra-multi-scale-implicit-decomposition-for-deepfake-detection", "canonical_source": "https://arxiv.org/abs/2610.31979", "published_at": "2026-10-26 04:00:00+00:00", "updated_at": "2026-10-26 04:12:23.146624+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "computer-vision", "ai-research", "ai-safety"], "entities": ["Spectra", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/spectra-multi-scale-implicit-decomposition-for-deepfake-detection", "markdown": "https://wpnews.pro/news/spectra-multi-scale-implicit-decomposition-for-deepfake-detection.md", "text": "https://wpnews.pro/news/spectra-multi-scale-implicit-decomposition-for-deepfake-detection.txt", "jsonld": "https://wpnews.pro/news/spectra-multi-scale-implicit-decomposition-for-deepfake-detection.jsonld"}}