Steganalysis models excel on benchmark datasets but struggle in the wild when analyzed images are produced by a processing pipeline unseen during training. This problem known as Cover Source Mismatch (CSM) is particularly hard in realistic settings where practitioners (1) have access to only a small, unlabeled dataset, (2) are unsure of the processing techniques applied to these images, and (3) lack information on the proportion of covers and stegos in that set. To answer this challenge, we introduce TADA (Target Alignment through Data Adaptation), a framework learning to emulate the unknown processing pipeline from a small unlabeled target set. This architecture is trained with a loss combining residual covariance alignment, residual distribution matching, and a $\ell^2$ loss constraining the emulator to produce realistic images. Across toy and operational targets, TADA yields substantial gains in robustness to CSM and improves operational generalization compared to strong holistic and atomistic baselines. Additional resources are available at this link: https://github.com/RonyAbecidan/TADA

翻译：隐写分析模型在基准数据集上表现优异，但在实际场景中，当待分析图像由训练阶段未见的处理流水线生成时，其性能会显著下降。这种被称为"载体源失配"（CSM）的问题在现实环境中尤为棘手，因为实际使用者（1）仅能获取少量无标签数据集，（2）不确定这些图像所采用的处理技术，（3）缺乏该数据集中载体图像与隐写图像比例的先验信息。为应对这一挑战，我们提出TADA（通过数据自适应实现目标对齐）框架，该框架能够从少量无标签目标数据集中学习模拟未知处理流水线。该架构采用融合残差协方差对齐、残差分布匹配以及约束模拟器生成逼真图像的$\ell^2$范数损失的联合损失函数进行训练。在仿真目标与真实目标场景的实验中，与强基线整体模型和原子模型相比，TADA在CSM鲁棒性方面取得了显著提升，并改善了操作泛化性能。补充资源请访问：https://github.com/RonyAbecidan/TADA