D$^3$：通过从差异中学习扩展深度伪造检测规模
D$^3$: Scaling Up Deepfake Detection by Learning from Discrepancy
AI总结：
针对现有深度伪造检测方法在多生成器训练下泛化与域内性能难以兼顾的问题，提出D$^3$框架，通过并行分支引入差异信号解构通用伪影，实现域外准确率提升5.3%且保持域内性能。
摘要：
生成式AI的蓬勃发展带来了机遇，也夹杂着风险与担忧。现有研究强调深度伪造检测在未见生成器上的泛化能力，显著提升了检测器识别更通用伪影的能力。本工作致力于向具备更优泛化性与鲁棒性的通用深度伪造检测系统迈进一步。我们首先将现有检测任务的训练设置从单生成器扩展到多生成器，在此过程中揭示了先前方法设计中存在的两个挑战，并证明了检测器性能的差异性。具体而言，我们发现当前专为单一生成器训练定制的方法，要么难以从多生成器中学习全面的伪影，要么会牺牲其在已见生成器上的拟合能力（即域内（ID）性能），以换取在未见生成器上的泛化能力（即域外（OOD）性能）。为解决上述挑战，我们提出了差异深度伪造检测器（Discrepancy Deepfake Detector，D$^3$）框架，其核心思想是通过引入一个并行网络分支，将失真图像特征作为额外的差异信号并补充其原始对应特征，从而解构来自多生成器的通用伪影。大量扩展实验证明了D$^3$的有效性：与当前最优（SOTA）方法相比，其在域外测试中准确率提升了5.3%，同时保持了域内性能。源代码将在我们的GitHub仓库更新：https://github.com/BigAandSmallq/D3
英文摘要：
The boom of Generative AI brings opportunities entangled with risks and concerns. Existing literature emphasizes the generalization capability of deepfake detection on unseen generators, significantly promoting the detector's ability to identify more universal artifacts. This work seeks a step toward a universal deepfake detection system with better generalization and robustness. We do so by first scaling up the existing detection task setup from the one-generator to multiple-generators in training, during which we disclose two challenges presented in prior methodological designs and demonstrate the divergence of detectors' performance. Specifically, we reveal that the current methods tailored for training on one specific generator either struggle to learn comprehensive artifacts from multiple generators or sacrifice their fitting ability for seen generators (i.e., In-Domain (ID) performance) to exchange the generalization for unseen generators (i.e., Out-Of-Domain (OOD) performance). To tackle the above challenges, we propose our Discrepancy Deepfake Detector (D$^3$) framework, whose core idea is to deconstruct the universal artifacts from multiple generators by introducing a parallel network branch that takes a distorted image feature as an extra discrepancy signal and supplement its original counterpart. Extensive scaled-up experiments demonstrate the effectiveness of D$^3$, achieving 5.3% accuracy improvement in the OOD testing compared to the current SOTA methods while maintaining the ID performance. The source code will be updated in our GitHub repository: https://github.com/BigAandSmallq/D3
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
