DNA：用于生成图像源追踪的双阶段原生归因
DNA: Dual-stage Native Attribution for Generated Image Source Tracing
- University of Science and Technology of China（中国科学技术大学）
- Hefei University of Technology（合肥工业大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对图像生成中家族内变体源追踪难题，提出双阶段原生归因（DNA）框架。粗粒度用 AEDR 筛选，细粒度用 NPC 归因。构建 DNA-30K 基准，实验表明 DNA 准确率高，优于基线，为图像源追踪提供有效方法。
AI中文摘要：
图像生成的快速发展产生了大量家族内变体，使得可疑图像的源模型归因在数字取证中变得越发重要。现有主动方法依赖水印嵌入或模型修改，会降低视觉质量并限制部署灵活性。被动方法常依赖大规模监督训练或单一重建信号，限制了处理未知源和区分家族内高度相似变体的能力。我们发现潜在生成模型中的归因信号在架构层次上自然分层：VAE 级线索反映家族共享信息，而主干级线索捕获变体特定行为。基于此，我们提出双阶段原生归因（DNA），这是一个从粗到细的框架，无需额外神经网络训练即可遵循此层次结构。粗粒度阶段使用自动编码器双重重建（AEDR）进行高效的开放集家族级筛选。细粒度阶段使用原生预测一致性（NPC）进行封闭集模型级归因，它在语义条件下比较多个噪声水平下家族内变体的原生预测误差，并通过归一化校准分数归因源。为了进行系统评估，我们构建了 DNA-30K，这是一个用于开放集家族级评估下家族内变体归因的基准。它包括由六个家族的 24 个候选模型生成的 30000 张图像，涵盖去噪扩散和流匹配，以及非候选生成图像和自然图像作为未知源。实验表明，在随机猜测准确率低于 1%的任务上，DNA 实现了 89.11%的端到端归因准确率，即使将 AEDR 用作粗粒度阶段，也比最强基线高出 33.81%。
英文摘要：
The rapid evolution of image generation has produced numerous within-family variants, making source-model attribution of suspect images increasingly important for digital forensics. Existing proactive methods rely on watermark embedding or model modification, which may degrade visual quality and limit deployment flexibility. Passive methods often rely on large-scale supervised training or a single reconstruction signal, limiting their ability to handle unknown sources and distinguish highly similar within-family variants. We observe that attribution signals in latent generative models are naturally stratified across architectural levels: VAE-level cues reflect family-shared information, whereas backbone-level cues capture variant-specific behaviors. Motivated by this insight, we propose Dual-stage Native Attribution (DNA), a coarse-to-fine framework that follows this hierarchy without additional neural-network training. The coarse-grained stage uses Autoencoder Double-Reconstruction (AEDR) for efficient open-set family-level screening. The fine-grained stage performs closed-set model-level attribution with Native Prediction Consistency (NPC), which compares native prediction errors of within-family variants across multiple noise levels under semantic conditioning and attributes the source via normalized calibrated scores. To enable systematic evaluation, we construct DNA-30K, a benchmark for within-family variant attribution under open-set family-level evaluation. It comprises 30,000 images generated by 24 candidate models across six families spanning both denoising diffusion and flow matching, plus non-candidate generated and natural images as unknown sources. Experiments show that DNA achieves 89.11% end-to-end attribution accuracy on a task where random guessing accuracy is below 1% and outperforms the strongest baseline by 33.81% even when AEDR is used as the coarse-grained stage.
