analiza aiqbench
This research presents a dual-branch ensemble framework that combines semantic deep learning with mathematical forensic features to accurately identify the source of synthetic images, even when they have undergone post-processing like JPEG compression. By leveraging spectral profiles and noise residuals analyzed on standard CPUs without GPU acceleration, the method offers a highly efficient solution for detecting distribution shifts in real-world deployed text-to-image models. The approach achieves 95.60% accuracy across ten different generators, demonstrating significant practical utility for content provenance verification in an era of advanced generative AI.
Hybrid Semantic and Spectral Ensemble for Robust Synthetic Image Source Attribution
arXiv:2607.22808v1 Announce Type: cross Abstract: The rapid advancement of text-to-image (T2I) models has necessitated robust Synthetic Image Source Attribution (SIA) methodologies. A critical challenge in SIA is the distribution shift between pristine training images and real-world deployed images, which undergo unknown post-processing operations such as JPEG compression and blurring. In this work, proposed for the DLMMDD Challenge at ICANN 2026, we introduce a dual-branch ensemble framework fusing Semantic Deep Learning with Mathematical Forensic Feature Extraction. The semantic branch employs EfficientNet-B0 regularized with Exponential Moving Averaging (EMA) and Label Smoothing. The forensic branch extracts 126 mathematical features -- including SVD spectral profiles and Local Binary Patterns -- from high-pass noise residuals, compressed via Truncated SVD and classified with XGBoost. Evaluated on a dataset of 10 generators where 55% of the test set is degraded, our approach achieves a private leaderboard accuracy of 95.60%. Furthermore, the entire pipeline is highly computationally efficient, requiring no GPU acceleration and executing end-to-end on a standard CPU in under 6.5 hours, highlighting the practicality and scalability of mathematical forensics for real-world deployment.
Abstract:The rapid advancement of text-to-image (T2I) models has necessitated robust Synthetic Image Source Attribution (SIA) methodologies. A critical challenge in SIA is the distribution shift between pristine training images and real-world deployed images, which undergo unknown post-processing operations such as JPEG compression and blurring. In this work, proposed for the DLMMDD Challenge at ICANN 2026, we introduce a dual-branch ensemble framework fusing Semantic Deep Learning with Mathematical Forensic Feature Extraction. The semantic branch employs EfficientNet-B0 regularized with Exponential Moving Averaging (EMA) and Label Smoothing. The forensic branch extracts 126 mathematical features -- including SVD spectral profiles and Local Binary Patterns -- from high-pass noise residuals, compressed via Truncated SVD and classified with XGBoost. Evaluated on a dataset of 10 generators where 55% of the test set is degraded, our approach achieves a private leaderboard accuracy of 95.60%. Furthermore, the entire pipeline is highly computationally efficient, requiring no GPU acceleration and executing end-to-end on a standard CPU in under 6.5 hours, highlighting the practicality and scalability of mathematical forensics for real-world deployment.
Peer-reviewed and accepted to the DLMMDD Challenge Workshop at the 35th International Conference on Artificial Neural Networks (ICANN 2026). 7 pages, 1 figures
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Graphics (cs.GR); Image and Video Processing (eess.IV)
arXiv:2607.22808 [cs.CV]
(or arXiv:2607.22808v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2607.22808 arXiv-issued DOI via DataCite (pending registration)
From: Md. Ajwad Hossain [view email] [v1] Fri, 24 Jul 2026 17:22:33 UTC (25 KB)
Text integral sub licența deschisă a publicației · 240 cuvinte
Subiecte conexe
[AINews] OpenAI reports Navier-Stokes singularity find in 88 hours using Astra-next, roughly 10,000 agents and 130B tokens (>$40M), a contender for second ever Millennium Prize awarded
Latent Space · 9 sept. 2026
SynDORBench: Evaluating LVLM Perceptual Robustness Under Physically Constrained Visibility Conditions
arXiv · 29 sept. 2026
[AINews] Claude Opus 5.5, the new default model for AINews — and everybody cuts prices 40-50%
Latent Space · 23 sept. 2026
Article: The AI Productivity Paradox in Test Automation: Moving beyond Structural Validation to Perception and Intent
InfoQ · 1 iun. 2026