This work explores knowledge distillation (KD) for visually-rich document (VRD) applications such as document layout analysis (DLA) and document image classification (DIC). While VRD research is dependent on increasingly sophisticated and cumbersome models, the field has neglected to study efficiency via model compression. Here, we design a KD experimentation methodology for more lean, performant models on document understanding (DU) tasks that are integral within larger task pipelines. We carefully selected KD strategies (response-based, feature-based) for distilling knowledge to and from backbones with different architectures (ResNet, ViT, DiT) and capacities (base, small, tiny). We study what affects the teacher-student knowledge gap and find that some methods (tuned vanilla KD, MSE, SimKD with an apt projector) can consistently outperform supervised student training. Furthermore, we design downstream task setups to evaluate covariate shift and the robustness of distilled DLA models on zero-shot layout-aware document visual question answering (DocVQA). DLA-KD experiments result in a large mAP knowledge gap, which unpredictably translates to downstream robustness, accentuating the need to further explore how to efficiently obtain more semantic document layout awareness.

针对视觉丰富的文档应用（如文档布局分析和文档图像分类），本文探讨了知识蒸馏（KD）。通过设计一种KD实验方法，我们研究了不同架构和容量的骨干模型之间的知识传递策略对教师-学生知识差距的影响，并发现一些方法可以始终优于监督学生训练。此外，我们设计了下游任务设置，评估了蒸馏的布局分析模型在零样本布局感知文档视觉问答上的鲁棒性，结果表明存在较大的知识差距，强调进一步探索如何高效获得更多的语义文档布局意识的必要性。

DistilDoc: 视觉内容丰富文档应用的知识蒸馏