mlsanityData to 14 Oct 2026

DTrOCR: Decoder-only Transformer for Optical Character Recognition

arXiv 2308.15996 ↗First posted 30 Aug 2023cs.CVcited by 40

Not on the map. Why?

Abstract (arXiv)

Typical text recognition methods rely on an encoder-decoder structure, in which the encoder extracts features from an image, and the decoder produces recognized text from these features. In this study, we propose a simpler and more effective method for text recognition, known as the Decoder-only Transformer for Optical Character Recognition (DTrOCR). This method uses a decoder-only Transformer to take advantage of a generative language model that is pre-trained on a large corpus. We examined whether a generative language model that has been successful in natural language processing can also be effective for text recognition in computer vision. Our experiments demonstrated that DTrOCR outperforms current state-of-the-art methods by a large margin in the recognition of printed, handwritten, and scene text in both English and Chinese.

Key facts. Cited by 40 papers in the map (data through 14 Oct 2026). Builds on 47 of its 68 references that are in the map. First posted on arXiv 30 Aug 2023.

Builds on

47 of 68 references.

References · 47

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
  5. 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,918
  6. 2020The Pile: An 800GB Dataset of Diverse Text for Language Modelingcited by 2,411
  7. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,269
  8. 2022LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scalecited by 906
  9. 2019LayoutLM: Pre-training of Text and Layout for Document Image Understandingcited by 553
  10. 2019CCNet: Extracting High Quality Monolingual Datasets from Web Crawl Datacited by 548
  11. 2015An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognitioncited by 520
  12. 2016Synthetic Data for Text Localisation in Natural Imagescited by 396
35 more
  1. 2021TrOCR: Transformer-based Optical Character Recognition with Pre-trained Modelscited by 279
  2. 2021ICDAR2019 Competition on Scanned Receipt OCR and Information Extractioncited by 256
  3. 2016COCO-Text: Dataset and Benchmark for Text Detection and Recognition in Natural Imagescited by 236
  4. 2021Read Like Humans: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Recognitioncited by 194
  5. 2022Scene Text Recognition with Permuted Autoregressive Sequence Modelscited by 149
  6. 2019ICDAR2019 Robust Reading Challenge on Arbitrary-Shaped Text (RRC-ArT)cited by 147
  7. 2019ICDAR2019 Robust Reading Challenge on Multi-lingual Scene Text Detection and Recognition -- RRC-MLT-2019cited by 139
  8. 2020Towards Accurate Scene Text Recognition with Semantic Reasoning Networkscited by 131
  9. 2018Show, Attend and Read: A Simple and Strong Baseline for Irregular Text Recognitioncited by 130
  10. 2019ICDAR 2019 Robust Reading Challenge on Reading Chinese Text on Signboardcited by 128
  11. 2014Reading Text in the Wild with Convolutional Neural Networkscited by 122
  12. 2019ICDAR 2019 Competition on Large-scale Street View Text with Partial Labeling -- RRC-LSVTcited by 122
  13. 2020SEED: Semantics Enhanced Encoder-Decoder Framework for Scene Text Recognitioncited by 119
  14. 2017ICDAR2017 Competition on Reading Chinese Text in the Wild (RCTW-17)cited by 105
  15. 2019Decoupled Attention Network for Text Recognitioncited by 102
  16. 2021From Two to One: A New Scene Text Recognizer with Visual Language Modeling Networkcited by 97
  17. 2022SVTR: Scene Text Recognition with a Single Visual Modelcited by 88
  18. 2021Vision Transformer for Fast and Efficient Scene Text Recognitioncited by 81
  19. 2020Pay Attention to What You Read: Non-recurrent Handwritten Text-Line Recognitioncited by 71
  20. 2019TextScanner: Reading Characters in Order for Robust Scene Text Recognitioncited by 68
  21. 2019MASTER: Multi-Aspect Non-local Network for Scene Text Recognitioncited by 65
  22. 2021What If We Only Use Real Datasets for Scene Text Recognition? Toward Scene Text Recognition With Fewer Labelscited by 63
  23. 2019UER: An Open-Source Toolkit for Pre-training Modelscited by 62
  24. 2019Evaluating Sequence-to-Sequence Models for Handwritten Text Recognitioncited by 58
  25. 2020Handwritten Optical Character Recognition (OCR): A Comprehensive Systematic Literature Review (SLR)cited by 56
  26. 2021Benchmarking Chinese Text Recognition: Datasets, Baselines, and an Empirical Studycited by 53
  27. 2021Rethinking Text Line Recognition Modelscited by 45
  28. 2021SynthTIGER: Synthetic Text Image GEneratoR Towards Better Text Recognition Modelscited by 43
  29. 2022MaskOCR: Text Recognition with Masked Encoder-Decoder Pretrainingcited by 34
  30. 2019A Multi-Object Rectified Attention Network for Scene Text Recognitioncited by 30
  31. 2021Primitive Representation Learning for Scene Text Recognitioncited by 28
  32. 2021Joint Visual Semantic Reasoning: Multi-Stage Decoder for Text Recognitioncited by 24
  33. 2021Towards the Unseen: Iterative Text Recognition by Distilling from Errorscited by 12
  34. 2023DiffusionSTR: Diffusion Model for Scene Text Recognitioncited by 7
  35. 2023A3S: Adversarial learning of semantic representations for Scene-Text Spottingcited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 2; 2024: 11; 2025: 16; 2026: 11.

40 papers.

Sort:

By area:

  1. 2025TopV: Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision Language Modelcited by 57
  2. 2024DocVLM: Make Your VLM an Efficient Readercited by 28
  3. 2025Handwritten Text Recognition: A Surveycited by 12
  4. 2024On the Generalization of Handwritten Text Recognition Modelscited by 10
  5. 2024HATFormer: Historic Handwritten Arabic Text Recognition with Transformerscited by 9
  6. 2024DLaVA: Document Language and Vision Assistant for Answer Localization with Enhanced Interpretability and Trustworthinesscited by 8
  7. 2024TextBlockV2: Towards Precise-Detection-Free Scene Text Spotting with Pre-trained Language Modelcited by 5
  8. 2023RL-LOGO: Deep Reinforcement Learning Localization for Logo Recognitioncited by 4
  9. 2024Mixed Text Recognition with Efficient Parameter Fine-Tuning and Transformercited by 4
  10. 2024General Detection-based Text Line Recognitioncited by 4
  11. 2023TPPoet: Transformer-Based Persian Poem Generation using Minimal Data and Advanced Decoding Techniquescited by 2
  12. 2025InkFM: A Foundational Model for Full-Page Online Handwritten Note Understandingcited by 2
28 more
  1. 2025MetaWriter: Personalized Handwritten Text Recognition Using Meta-Learned Prompt Tuningcited by 2
  2. 2024JSTR: Judgment Improves Scene Text Recognitioncited by 1
  3. 2024Spatial Context-based Self-Supervised Learning for Handwritten Text Recognitioncited by 1
  4. 2024JaPOC: Japanese Post-OCR Correction Benchmark using Voucherscited by 1
  5. 2025Judge a Book by its Cover: Investigating Multi-Modal LLMs for Multi-Page Handwritten Document Transcriptioncited by 1
  6. 2025Practical Fine-Tuning of Autoregressive Models on Limited Handwritten Textscited by 1
  7. 2025Leveraging Contrast Information for Efficient Document Shadow Removalcited by 1
  8. 2025HTR-ConvText: Leveraging Convolution and Textual Information for Handwritten Text Recognitioncited by 1
  9. 2026DRetHTR: Linear-Time Decoder-Only Retentive Network for Handwritten Text Recognitioncited by 1
  10. 2026Few-shot Writer Adaptation via Multimodal In-Context Learningcited by 1
  11. 2026Democratizing the medieval English legal traditioncited by 1
  12. 2026EpiSAM: Character Segmentation in Challenging Stone Inscriptionscited by 1
  13. 2024Segmentation-free Connectionist Temporal Classification loss based OCR Model for Text Captcha Classificationcited by 0
  14. 2025Wormhole Memory: A Rubik's Cube for Cross-Dialogue Retrievalcited by 0
  15. 2025Accurate Scene Text Recognition with Efficient Model Scaling and Cloze Self-Distillationcited by 0
  16. 2025Coarse-to-Fine Learning for Multi-Pipette Localisation in Robot-Assisted In Vivo Patch-Clampcited by 0
  17. 2025A Lightweight Multi-Module Fusion Approach for Korean Character Recognitioncited by 0
  18. 2025Preserving Privacy Without Compromising Accuracy: Machine Unlearning for Handwritten Text Recognitioncited by 0
  19. 2025DOTA: Deformable Optimized Transformer Architecture for End-to-End Text Recognition with Retrieval-Augmented Generationcited by 0
  20. 2025AutoSign: Direct Pose-to-Text Translation for Continuous Sign Language Recognitioncited by 0
  21. 2025TEACH: Text Encoding as Curriculum Hints for Scene Text Recognitioncited by 0
  22. 2026Efficient Domain Adaptation for Text Line Recognition via Decoupled Language Modelscited by 0
  23. 2026Zero-Shot Synthetic-to-Real Handwritten Text Recognition via Task Analogiescited by 0
  24. 2026Adapting TrOCR for Printed Tigrinya Text Recognition: Word-Aware Loss Weighting for Cross-Script Transfer Learningcited by 0
  25. 2026Masked Next-Scale Prediction for Self-supervised Scene Text Recognitioncited by 0
  26. 2026ExpertHTR: Unified Handwritten Text Recognition with Multi-Task Learning and Sparse Mixture-of-Expertscited by 0
  27. 2026A Free Lunch? Adapting PP-OCRv6 for Historical Text Recognitioncited by 0
  28. 2026Handwritten Text Recognition Lives in the High-Pixel Variance Subspacecited by 0