PE-Mamba: Bidirectional Selective Layer Aggregation for AI-Generated Image Detection

• URL: http://arxiv.org/abs/2608.07999v1
• Date: Sat, 08 Aug 2026 08:24:44 GMT
• Title: PE-Mamba: Bidirectional Selective Layer Aggregation for AI-Generated Image Detection
• Abstract summary: textbfPE-Mamba is a novel framework built upon a pre-trained PE-Core vision transformer with lightweight LoRA adaptation.<n>A bidirectional selective aggregator (BSA) processes layer-wise classification tokens through forward and backward selective scans.<n>A softmax-weighted aggregator (SWA) computes a learned global summary of all layer tokens as a complementary aggregation path.<n>A sigmoid-gated blend (SGA) adaptively fuses the BSA and SWA outputs via a learnable scalar gate.
• Score: 2.1410799064827235
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: AI-generated image (AIGI) detection has become increasingly challenging due to the rapid advancement of generative models and the diminishing gap between synthetic and authentic content. Existing vision transformer-based detectors commonly rely on weighted-sum strategies to aggregate intermediate representations across transformer layers, often overlooking the inherently ordered semantic progression of hierarchical features from shallow texture cues to deep semantic representations. In this work, we propose \textbf{PE-Mamba}, a novel framework built upon a pre-trained PE-Core vision transformer with lightweight LoRA adaptation that introduces three complementary components for cross-layer feature aggregation and fusion. First, a bidirectional selective aggregator (BSA) processes layer-wise classification tokens through forward and backward selective scans, where the forward scan progressively accumulates shallow-to-deep forensic evidence, and the backward scan performs deep-to-shallow contextual refinement to reinterpret low-level cues in light of high-level semantic context. Second, a softmax-weighted aggregator (SWA) computes a learned global summary of all layer tokens as a complementary aggregation path. Third, a sigmoid-gated blend (SGA) adaptively fuses the BSA and SWA outputs via a learnable scalar gate, allowing the model to dynamically balance directional sequential evidence and global layer-wise aggregation. Extensive experiments on UniversalFakeDetect (96.6\% mACC, 99.5\% mAP) and AIGCDetect (95.3\% mACC, 98.1\% mAP) demonstrate that \methodname{} outperforms 18 detectors with superior generalization across diverse generative models, while training only 1.3\% of total parameters (0.13\% for LoRA alone).

Related papers

• HEDGE: Heterogeneous Ensemble for Detection of AI-GEnerated Images in the Wild [17.30735058274559]
  HEDGE is a Heterogeneous Ensemble for Detection of AI-GEnerated images.<n> RouteA progressively constructs DINOv3-based detectors through staged data expansion and augmentation escalation.<n> RouteB incorporates a higher-resolution branch for fine-grained forensic cues, and RouteC adds a MetaCLIP2-based branch for backbone diversity.
  arXiv  Detail & Related papers  (2026-04-04T02:46:36Z)
• Enhancing Gradient Inversion Attacks in Federated Learning via Hierarchical Feature Optimization [56.95448807869383]
  Federated Learning (FL) has emerged as a compelling paradigm for privacy-preserving distributed machine learning.<n>Recent studies find that the gradients exchanged in the FL system are also vulnerable to privacy leakage.<n>We propose textbfGradient textbfInversion over textbfFeature textbfDomains (GIFD)
  arXiv  Detail & Related papers  (2026-04-01T14:32:15Z)
• Multi-Feature Fusion Approach for Generative AI Images Detection [3.16770435670322]
  We investigate and evaluate a multi-feature fusion framework that combines complementary cues from three distinct spaces.<n>We show that individual feature spaces exhibit significant performance variability across different generators.<n>Compared to state-of-the-art methods, the proposed framework yields consistently improved performance across all evaluated datasets.
  arXiv  Detail & Related papers  (2026-03-31T14:25:10Z)
• LGEST: Dynamic Spatial-Spectral Expert Routing for Hyperspectral Image Classification [2.866506214681715]
  Local-Global Expert Spatial-Spectral Transformer (LGEST) is a novel framework that combines three key innovations.<n>The LGEST consistently outperforms state-of-the-art methods in experiments on four benchmark datasets.
  arXiv  Detail & Related papers  (2026-03-25T07:53:31Z)
• When Detectors Forget Forensics: Blocking Semantic Shortcuts for Generalizable AI-Generated Image Detection [48.75021863792367]
  Vision Foundation Models (VFMs, emphe.g., CLIP) often struggle to generalize to images created using unseen generation pipelines.<n>We propose textbfGeometric Semantic Decoupling (GSD), a parameter-free module that explicitly removes semantic components from learned representations.<n>GSD estimates semantic directions from batch-wise statistics and projects them out via a geometric constraint, forcing the artifact detector to rely on semantic-invariant forensic evidence.
  arXiv  Detail & Related papers  (2026-03-10T06:16:35Z)
• CtrlFuse: Mask-Prompt Guided Controllable Infrared and Visible Image Fusion [51.060328159429154]
  Infrared and visible image fusion generates all-weather perception-capable images by combining complementary modalities.<n>We propose CtrlFuse, a controllable image fusion framework that enables interactive dynamic fusion guided by mask prompts.<n> Experiments demonstrate state-of-the-art results in both fusion controllability and segmentation accuracy, with the adapted task branch even outperforming the original segmentation model.
  arXiv  Detail & Related papers  (2026-01-12T13:36:48Z)
• Improving Deepfake Detection with Reinforcement Learning-Based Adaptive Data Augmentation [60.04281435591454]
  CRDA (Curriculum Reinforcement-Learning Data Augmentation) is a novel framework guiding detectors to progressively master multi-domain forgery features.<n>Central to our approach is integrating reinforcement learning and causal inference.<n>Our method significantly improves detector generalizability, outperforming SOTA methods across multiple cross-domain datasets.
  arXiv  Detail & Related papers  (2025-11-10T12:45:52Z)
• CINEMAE: Leveraging Frozen Masked Autoencoders for Cross-Generator AI Image Detection [25.84217122259626]
  CINEMAE adapts the core principles of text detection methods to the visual domain.<n>We trained exclusively on Stable Diffusion v1.4, our method achieves over 95% accuracy on all eight unseen generators in the GenImage benchmark.<n>This demonstrates that context-conditional reconstruction uncertainty provides a robust, transferable signal for AIGC detection.
  arXiv  Detail & Related papers  (2025-11-09T11:05:45Z)
• Hierarchical Fusion and Joint Aggregation: A Multi-Level Feature Representation Method for AIGC Image Quality Assessment [0.9821874476902972]
  The quality assessment of AI-generated content (AIGC) faces multi-dimensional challenges, that span from low-level visual perception to high-level semantic understanding.<n>To address this limitation, a multi-level visual representation paradigm is proposed with three stages, namely multi-level feature extraction, hierarchical fusion, and joint aggregation.<n>Experiments on benchmarks demonstrate outstanding performance on both tasks, validating the effectiveness of the proposed multi-level visual assessment paradigm.
  arXiv  Detail & Related papers  (2025-07-23T04:12:32Z)
• S^2Former-OR: Single-Stage Bi-Modal Transformer for Scene Graph Generation in OR [50.435592120607815]
  Scene graph generation (SGG) of surgical procedures is crucial in enhancing holistically cognitive intelligence in the operating room (OR) Previous works have primarily relied on multi-stage learning, where the generated semantic scene graphs depend on intermediate processes with pose estimation and object detection. In this study, we introduce a novel single-stage bi-modal transformer framework for SGG in the OR, termed S2Former-OR.
  arXiv  Detail & Related papers  (2024-02-22T11:40:49Z)
• GSMFlow: Generation Shifts Mitigating Flow for Generalized Zero-Shot Learning [55.79997930181418]
  Generalized Zero-Shot Learning aims to recognize images from both the seen and unseen classes by transferring semantic knowledge from seen to unseen classes. It is a promising solution to take the advantage of generative models to hallucinate realistic unseen samples based on the knowledge learned from the seen classes. We propose a novel flow-based generative framework that consists of multiple conditional affine coupling layers for learning unseen data generation.
  arXiv  Detail & Related papers  (2022-07-05T04:04:37Z)

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.