For visual document understanding (VDU), self-supervised pretraining has been shown to successfully generate transferable representations, yet, effective adaptation of such representations to distribution shifts at test-time remains to be an unexplored area. We propose DocTTA, a novel test-time adaptation method for documents, that does source-free domain adaptation using unlabeled target document data. DocTTA leverages cross-modality self-supervised learning via masked visual language modeling, as well as pseudo labeling to adapt models learned on a \textit{source} domain to an unlabeled \textit{target} domain at test time. We introduce new benchmarks using existing public datasets for various VDU tasks, including entity recognition, key-value extraction, and document visual question answering. DocTTA shows significant improvements on these compared to the source model performance, up to 1.89\% in (F1 score), 3.43\% (F1 score), and 17.68\% (ANLS score), respectively. Our benchmark datasets are available at \url{https://saynaebrahimi.github.io/DocTTA.html}.

翻译：针对视觉文档理解（VDU），自监督预训练已被证明能够有效生成可迁移的表征，然而在测试时如何使此类表征有效适应分布偏移仍是一个未被探索的领域。我们提出DocTTA——一种面向文档的新型测试时自适应方法，该方法利用无标签的目标文档数据进行无源域适配。DocTTA通过基于掩码视觉语言建模的跨模态自监督学习，以及伪标签技术，在测试时将\textit{源域}学到的模型适配至无标签的\textit{目标域}。我们利用现有公开数据集为多种VDU任务（包括实体识别、键值抽取和文档视觉问答）构建了新的基准测试。与源域模型性能相比，DocTTA在各项任务上分别取得了高达1.89%（F1分数）、3.43%（F1分数）和17.68%（ANLS分数）的显著提升。我们的基准数据集可在\url{https://saynaebrahimi.github.io/DocTTA.html}获取。