Detecting Diffusion-generated Images via Dynamic Assembly ForestsDetecting Diffusion-generated Images via Dynamic Assembly Forests

2026-04-10

AI Summary

This work addresses the pressing need for efficient and reliable detection methods to mitigate security risks posed by images generated via diffusion models. We propose Dynamic Assembly Forest (DAF), a lightweight detector based on the deep forest paradigm, which introduces an enhanced deep forest architecture to the task of synthetic image detection for the first time. DAF integrates multi-granularity feature learning with a scalable training mechanism, enabling highly efficient inference without reliance on GPU acceleration. Experimental results demonstrate that DAF achieves detection performance comparable to state-of-the-art deep neural networks on standard benchmarks, while substantially reducing model parameters and computational overhead, thereby offering a practical solution for resource-constrained environments.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Neural Architectures and Foundation Models
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Diffusion models are known for generating high-quality images, causing serious security concerns. To combat this, most efforts rely on deep neural networks (e.g., CNNs and Transformers), while largely overlooking the potential of traditional machine learning models. In this paper, we freshly investigate such alternatives and proposes a novel Dynamic Assembly Forest model (DAF) to detect diffusion-generated images. Built upon the deep forest paradigm, DAF addresses the inherent limitations in feature learning and scalable training, making it an effective diffusion-generated image detector. Compared to existing DNN-based methods, DAF has significantly fewer parameters, much lower computational cost, and can be deployed without GPUs, while achieving competitive performance under standard evaluation protocols. These results highlight the strong potential of the proposed method as a practical substitute for heavyweight DNN models in resource-constrained scenarios. Our code and models are available at https://github.com/OUC-VAS/DAF.

Problem

• diffusion-generated images
• image detection
• deep learning alternatives
• security concerns
• machine learning

Innovation

• Dynamic Assembly Forest
• diffusion-generated image detection
• deep forest
• lightweight model
• traditional machine learning