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.
arXiv:2610.01778v1 Announce Type: new Abstract: Reliable evaluation of image forgery localization (IFL) requires assessing models under diverse distribution changes, yet existing benchmarks often cov...
arXiv:2608.20929v1 Announce Type: new Abstract: AI-generated image manipulation localization identifies edited pixels, but its OOD performance lags behind image-level detection partly because pixel s...
The paper proposes a unified forensics framework that extends traditional binary image manipulation detection to a multiclass setting—distinguishing real, fully synthetic, and tampered images. It adds a segmentation branch for pixel‑level localization of tampered regions, achieving higher classification accuracy and IoU scores compared to recent benchmarks. The authors provide the implementation on GitHub for reproducibility.
arXiv:2610.08639v1 Announce Type: new Abstract: As generated images become increasingly realistic, reliable forgery detection is essential for maintaining trust in visual information. However, existi...
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.
arXiv:2609.36882v1 Announce Type: new Abstract: Localizing AI-edited regions is essential for interpretable forensic analysis, but remains challenging due to subtle and spatially distributed artifact...
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...
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:2609.38251v1 Announce Type: cross Abstract: The rapid evolution of image manipulation techniques has raised growing public security concerns. Existing Image Forgery Localization (IFL) methods c...
The paper proposes Artifact-Complementary Expert Fusion (ACEF), a two‑stage framework that enhances AI‑generated image detection by combining two types of reconstruction artifacts—VAE/DDIM and SRGAN—into aligned synthetic negatives. ACEF first builds artifact‑specific experts using LoRA adaptation on a frozen backbone, then fuses their multi‑layer evidence with Layer‑wise Artifact‑Complementary Fusion (LACF) to mitigate conflicts between artifact manifolds. Experiments on 13 benchmarks show that this approach improves generalizability over existing state‑of‑the‑art methods.
arXiv:2610.01544v1 Announce Type: new Abstract: Existing image forgery detectors often suffer from generalization to unseen manipulation methods due to the limited ability to capture transferable for...
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.