发表机构
Sakana AI（Sakana AI）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对AI生成图像检测在多因素变化下失效的问题，提出先验条件高斯判别梯方法，在Percept-Lens数据集上验证其性能，推动相关报告与基线的优化。
AI 中文摘要
基于扩散模型的生成器使合成图像变得无处不在，但检测器在生成器、提示/风格和源域同时发生变化时往往失效。我们将AI生成图像检测作为一个由训练先验、冻结编码器特征空间和决策规则描述的迁移系统进行研究，探究分类器头训练在现代特征已可区分的情况下是否仍有价值。作为受控诊断方法，我们拟合了一个先验条件高斯判别梯：基于一阶和二阶特征统计量、在嵌套协方差假设下构建的闭式解分类头。在覆盖39个公开数据集（共710万张图像）的统一协议Percept-Lens上，当匹配先验和编码器时，最优梯级的性能常与已发布的AI生成图像检测器头相当，有时甚至更优。我们进一步量化了其对训练先验的强敏感性、基于矩的分类头的数据效率，以及高斯偏移度量的表示依赖性，这推动了（先验、编码器、头）层面的报告，以及为AI生成图像迁移任务构建更强大的分析基线。
英文摘要
Diffusion-based generators have made synthetic images ubiquitous, but detectors often fail under simultaneous shifts in generator, prompt/style, and source-domain. We study AI-generated image detection as a transfer system described by training prior, frozen encoder feature space, and decision rule, and ask when classifier head training adds value beyond what is already separable in modern features. As a controlled diagnostic, we fit a prior-conditioned Gaussian discriminant ladder: closed-form heads built from first- and second-order feature statistics under nested covariance assumptions. On Percept-Lens, a unified protocol over 39 public datasets (7.1 million images), the best rung is frequently competitive with, and sometimes exceeds, released AI-generated image detector heads when matched on both prior and encoder. We further quantify strong sensitivity to the training prior, data-efficiency of moment-based heads, and representation dependence of Gaussian shift metrics, motivating (prior, encoder, head)-level reporting and stronger analytical baselines for AIGI transfer.
CommentsAccepted in ECCV 2026
