arXiv Computer Vision

Forensic Twins: Self-Supervised Residual Learning for AI-Generated Image Forensics

Forensic Twins introduces a Self‑Supervised Residual Learning (SSRL) framework that trains on real images only, using a frozen forensic residual extractor to generate two disjoint crops per image. The pretext task suppresses semantic content, focusing the model on the stationary fingerprint of the image acquisition pipeline, and achieves 56.61% accuracy in attributing AI‑generator sources, outperforming prior zero‑shot methods. When combined with an offline Gaussian Mixture Model, the approach reaches 97.99% AUC across 27 unseen AI generators, including GANs, diffusion models, and commercial systems.

arXiv Computer Vision

2d ago

RED: Reconstruction Evolution Dynamics for Generalizable AI-Generated Image Detection

The paper introduces RED (Reconstruction Evolution Dynamics), a new framework for detecting AI-generated images that leverages the evolution of intermediate reconstruction stages rather than relying solely on static representations or endpoint discrepancies. RED uses a frozen multiscale VQ‑VAE and a frozen CLIP encoder to capture a reconstruction trajectory, then learns image‑adaptive stage weights from token negative log‑likelihoods provided by a frozen VAR model. Experiments on six benchmarks show RED achieves the highest average accuracy (92.5%) and precision (97.5%) among evaluated methods, and it remains robust to common image degradations.

By Wenpeng Mu, Junshan Jin, Tanfeng Sun, Xinghao Jiang, Qiang Xu

benchmarks

arXiv Computer Vision

Aug 31

FUSED: Forensic-Semantic Mixture-of-Experts for AI Inpainting Detection and Localization

FUSED is a new framework that jointly detects and localizes AI-generated inpainting by combining low-level forensic cues with high-level semantic features through a sparsely-gated Mixture-of-Experts architecture. It predicts both an image-level manipulation score and a pixel-level mask of the inpainted region. On the OpenSDID cross-generator benchmark, FUSED outperforms existing methods, especially on unseen generators, and transfers effectively to the AutoSplice and CocoGlide benchmarks, doubling localization performance.

By Anton Nuzhdin, Marcel Worring, Ivona Najdenkoska

diffusion robotics benchmarks

arXiv Machine Learning

Jul 16

Fully AI-Generated Image Detection: Definition, Recent Advances and Challenges

arXiv:2502. 19716v3 Announce Type: replace-cross Abstract: Recent advances in visual generative models have enabled the creation of highly realistic, fully AI-generated images without relying on real source content.

By Qijie Xu, Can Wang, Jiawei Chen, Siwei Lyu, Defang Chen

arXiv AI

Jun 26

Perception, Verdict, and Evolution: Hindsight-Driven Self-Refining Forensics Agent for AI-Generated Image Detection

arXiv:2606. 26552v1 Announce Type: cross Abstract: The rapid advancement of generative models presents a significant challenge to existing deepfake detection methods, particularly given the widespread dissemination of highly realistic AI-generated images.

By Yangjun Wu, Keyu Yan, Yu Liu, Jingren Zhou, Fei Huang, Rong Zhang, Zhou Zhao, Fei Wu

llms agents fine-tuning multimodal benchmarks

arXiv AI

Jun 16

Deep Residual Injection for Full-Spectrum Forensic Signal Perception in Multimodal Large Language Models

arXiv:2606. 15880v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) have been increasingly adopted in forensics for their robust semantic understanding.

By Kaiqing Lin, Zhiyuan Yan, Ruoxin Chen, Ke-Yue Zhang, Yue Zhou, Caiyong Piao, Bin Li, Taiping Yao, Bo Wang, Youchang Xiao, Shouhong Ding

llms fine-tuning multimodal benchmarks

arXiv AI

Sep 18

FORGE: Forensic Reasoning with Grounded Evidence

FORGE is a forensic deepfake analysis system that provides region‑grounded natural language explanations for image manipulations. It addresses the inductive bias mismatch of multimodal large language models by adding a Vision‑Only Model trained on dense patch prediction, allowing the language model to interleave tokens with preserved spatial correspondence. Across face‑manipulated and fully synthetic content, FORGE delivers fine‑grained attribute queries and outperforms in‑domain baselines, with region‑specific evaluation and human studies confirming explanation faithfulness.

By Rohit Kundu, Shan Jia, Vishal Mohanty, Athula Balachandran, Amit K. Roy-Chowdhury

llms robotics multimodal safety

arXiv Computer Vision

1d ago

Agentic Tool-Augmented Reasoning for Explainable Image Forgery Detection

arXiv:2609.39066v1 Announce Type: new Abstract: Conventional image forgery detection methods produce binary scores or pixel-level masks without interpretable evidence, while recent multimodal large l...

By Zhiya Tan, Jing Huang, Changtao Miao, Lin Tan, Xin Zhang, Weiwei Feng, Jianshu Li, Joey Tianyi Zhou

llms agents reinforcement-learning fine-tuning multimodal benchmarks

arXiv AI

Aug 10

Multi-Agent Forensic Reasoning for Generalizable Deepfake Video Detection

arXiv:2608. 06865v1 Announce Type: cross Abstract: The malicious use of generative artificial intelligence to create highly realistic deepfake videos raises serious ethical concerns and poses substantial challenges to AI safety.

By Xuechao Zou, Shun Zhang, Kai Li, Yi Zhou, Xinyu Sun, Yuhui Chen, Zhe Wu, Congyan Lang, Junliang Xing

llms agents multimodal benchmarks safety

arXiv Computer Vision

Aug 24

LoRC: Detecting AI-Generated Images via Low-Rank Collapse in Semantic Residuals

arXiv:2608.20882v1 Announce Type: new Abstract: Modern generators faithfully model macroscopic semantics, producing synthetic images that appear highly realistic. Consequently, decisive forensic cues...

By Haozhen Yan, Ruoxin Chen, Jiahui Zhan, Bo Wang, Youchang Xiao, Shouhong Ding, Liqing Zhang, Taiping Yao, Jianfu Zhang

benchmarks

arXiv Computer Vision

Sep 16

Unifying Semantic Priors and High-Frequency Traces: Enhancing V-JEPA with Mixture-of-Experts for Robust Synthetic Image Forensics

The paper introduces MoE-JEPA, a dual‑stream deepfake detection model that combines a V‑JEPA backbone with a Residual Mixture‑of‑Experts mechanism and a noise stream branch. It further incorporates a Gated Attention Multiple Instance Learning module to refine spatial semantic understanding. On the SID‑Set benchmark, MoE‑JEPA achieves a new state‑of‑the‑art accuracy of 95.54%, outperforming much larger models.

By Simone Teglia, Irene Amerini

llms rag benchmarks

arXiv Computer Vision

Aug 25

AdaptPrompt: Parameter-Efficient Adaptation of VLMs for Generalizable Deepfake Detection

arXiv:2512.17730v2 Announce Type: replace Abstract: Detectors of AI-generated images tend to inherit the biases of the data they are trained on: models fitted to GAN imagery learn to treat GAN-specif...

By Yichen Jiang, Mohammed Talha Alam, Sohail Ahmed Khan, Duc-Tien Dang-Nguyen, Fakhri Karray

llms diffusion fine-tuning benchmarks safety

arXiv Machine Learning

Aug 4

Generated Images Are Easier to Forget: A Machine Unlearning Perspective for Synthetic Image Detection

arXiv:2608. 00716v1 Announce Type: cross Abstract: Robust detection of generated images is critical to counter the misuse of generative models.

By Jun Nie, Yonggang Zhang, Tongliang Liu, Yiu-ming Cheung, Bo Han, Xinmei Tian

benchmarks

Related stories