# DocTTT: Test-Time Training for Handwritten Document Recognition Using Meta-Auxiliary Learning

arXiv: https://arxiv.org/abs/2501.12898 · HTML: https://mlsanity.com/paper/2501.12898/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 35 of its 61 references that are in the map.
- First posted on arXiv 22 Jan 2025.

## Builds on (most cited first, 30 of 35)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://mlsanity.com/paper/1703.03400/) (2017)
- [On First-Order Meta-Learning Algorithms](https://mlsanity.com/paper/1803.02999/) (2018)
- [Meta-SGD: Learning to Learn Quickly for Few-Shot Learning](https://mlsanity.com/paper/1707.09835/) (2017)
- [AdaViT: Adaptive Tokens for Efficient Vision Transformer](https://mlsanity.com/paper/2112.07658/) (2021)
- [Contrastive Test-Time Adaptation](https://mlsanity.com/paper/2204.10377/) (2022)
- [Decoupled Attention Network for Text Recognition](https://mlsanity.com/paper/1912.10205/) (2019)
- [Few-Shot Class-Incremental Learning via Entropy-Regularized Data-Free Replay](https://mlsanity.com/paper/2207.11213/) (2022)
- [Pay Attention to What You Read: Non-recurrent Handwritten Text-Line Recognition](https://mlsanity.com/paper/2005.13044/) (2020)
- [MT3: Meta Test-Time Training for Self-Supervised Test-Time Adaption](https://mlsanity.com/paper/2103.16201/) (2021)
- [DAN: a Segmentation-free Document Attention Network for Handwritten Document Recognition](https://mlsanity.com/paper/2203.12273/) (2022)
- [End-to-end Handwritten Paragraph Text Recognition Using a Vertical Attention Network](https://mlsanity.com/paper/2012.03868/) (2020)
- [Meta-DMoE: Adapting to Domain Shift by Meta-Distillation from Mixture-of-Experts](https://mlsanity.com/paper/2210.03885/) (2022)
- [Improved Test-Time Adaptation for Domain Generalization](https://mlsanity.com/paper/2304.04494/) (2023)
- [MetaGCD: Learning to Continually Learn in Generalized Category Discovery](https://mlsanity.com/paper/2308.11063/) (2023)
- [MaskOCR: Text Recognition with Masked Encoder-Decoder Pretraining](https://mlsanity.com/paper/2206.00311/) (2022)
- [Test-Time Domain Adaptation by Learning Domain-Aware Batch Normalization](https://mlsanity.com/paper/2312.10165/) (2023)
- [Combinatorial Learning of Graph Edit Distance via Dynamic Embedding](https://mlsanity.com/paper/2011.15039/) (2020)
- [Full Page Handwriting Recognition via Image to Sequence Extraction](https://mlsanity.com/paper/2103.06450/) (2021)
- [Adapting to Distribution Shift by Visual Domain Prompt Generation](https://mlsanity.com/paper/2405.02797/) (2024)
- [Self-supervised Spatiotemporal Representation Learning by Exploiting Video Continuity](https://mlsanity.com/paper/2112.05883/) (2021)
- [SPAN: a Simple Predict & Align Network for Handwritten Paragraph Recognition](https://mlsanity.com/paper/2102.08742/) (2021)
- [DocSynth: A Layout Guided Approach for Controllable Document Image Synthesis](https://mlsanity.com/paper/2107.02638/) (2021)
- [Transformer-Based Approach for Joint Handwriting and Named Entity Recognition in Historical documents](https://mlsanity.com/paper/2112.04189/) (2021)
- [Boosting Modern and Historical Handwritten Text Recognition with Deformable Convolutions](https://mlsanity.com/paper/2208.08109/) (2022)
- [Distribution Alignment for Fully Test-Time Adaptation with Dynamic Online Data Streams](https://mlsanity.com/paper/2407.12128/) (2024)
- [HKR For Handwritten Kazakh & Russian Database](https://mlsanity.com/paper/2007.03579/) (2020)

## Built on by (most cited first, 3 of 3)

- [Few-shot Writer Adaptation via Multimodal In-Context Learning](https://mlsanity.com/paper/2603.29450/) (2026)
- [EVA-0: Test-Time Model Evolution with Only Two Forward Passes per Sample](https://mlsanity.com/paper/2605.18867/) (2026)
- [Hierarchical Adaptive networks with Task vectors for Test-Time Adaptation](https://mlsanity.com/paper/2508.09223/) (2025)

## Abstract (arXiv)

Despite recent significant advancements in Handwritten Document Recognition (HDR), the efficient and accurate recognition of text against complex backgrounds, diverse handwriting styles, and varying document layouts remains a practical challenge. Moreover, this issue is seldom addressed in academic research, particularly in scenarios with minimal annotated data available. In this paper, we introduce the DocTTT framework to address these challenges. The key innovation of our approach is that it uses test-time training to adapt the model to each specific input during testing. We propose a novel Meta-Auxiliary learning approach that combines Meta-learning and self-supervised Masked Autoencoder~(MAE). During testing, we adapt the visual representation parameters using a self-supervised MAE loss. During training, we learn the model parameters using a meta-learning framework, so that the model parameters are learned to adapt to a new input effectively. Experimental results show that our proposed method significantly outperforms existing state-of-the-art approaches on benchmark datasets.