| DTrOCR: Decoder-only Transformer for Optical Character Recognition | ✓ Link | 98.6 | DTrOCR 105M | 2023-08-30 |
| Multi-Granularity Prediction for Scene Text Recognition | ✓ Link | 98.3 | MGP-STR | 2022-09-08 |
| What Is Wrong with Synthetic Data for Scene Text Recognition? A Strong Synthetic Engine with Diverse Simulations and Self-Evolution | ✓ Link | 98.14 | MAERec + UnionST-S | 2026-02-06 |
| An Empirical Study of Scaling Law for OCR | ✓ Link | 98.13 | CLIP4STR-L* | 2023-12-29 |
| CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model | ✓ Link | 98.1 | CLIP4STR-L (DataComp-1B) | 2023-05-23 |
| CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model | ✓ Link | 97.4 | CLIP4STR-L | 2023-05-23 |
| CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model | ✓ Link | 97.2 | CLIP4STR-B | 2023-05-23 |
| Decoder Pre-Training with only Text for Scene Text Recognition | ✓ Link | 97.1 | PARSeq + DPTR (real-data training) (extra data) | 2024-08-11 |
| Instruction-Guided Scene Text Recognition | ✓ Link | 96.8 | IGTR-AR-PT (extra data) | 2024-01-31 |
| Multi-modal In-Context Learning Makes an Ego-evolving Scene Text Recognizer | ✓ Link | 96.74 | E²STR-ICL | 2023-11-22 |
| Context Perception Parallel Decoder for Scene Text Recognition | ✓ Link | 96.7 | CPPD | 2023-07-23 |
| Self-supervised Character-to-Character Distillation for Text Recognition | ✓ Link | 96.1 | CCD-ViT-Base | 2022-11-01 |
| OTE: Exploring Accurate Scene Text Recognition Using One Token | ✓ Link | 96.1 | OTE A/SVTR | 2024-06-17 |
| MonkeyOCRv2: A Visual-Text Foundation Model for Document AI | ✓ Link | 96.1 | PARSeq + MonkeyOCRv2-S | 2026-07-13 |
| Scene Text Recognition with Permuted Autoregressive Sequence Models | ✓ Link | 95.7±0.9 | PARSeq | 2022-07-14 |
| Embedding Rotation Invariance for Provable Multi-Oriented Scene Text Recognition | | 95.7 | RISTER-L (extra data) | 2026-08-11 |
| Advancing WordArt-Oriented Scene Text Recognition: Datasets and Methods | ✓ Link | 95.66 | WATERec-RS (extra data) | 2026-06-23 |
| Platypus: A Generalized Specialist Model for Reading Text in Various Forms | ✓ Link | 95.1 | Platypus (extra data) | 2024-08-27 |
| Revisiting Scene Text Recognition: A Data Perspective | ✓ Link | 94.4 | MAERec-B (Union14M-U PT + Union14M-L FT) | 2023-07-17 |
| Linguistics-aware Masked Image Modeling for Self-supervised Scene Text Recognition | | 94.0 | LMIM (Union14M-L fine-tuning) (extra data) | 2025-03-24 |
| SVTRv2: CTC Beats Encoder-Decoder Models in Scene Text Recognition | ✓ Link | 93.5 | SVTRv2-B | 2024-11-24 |
| TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models | ✓ Link | 93 | TrOCR Large (Syn+Benchmark) (extra data) | 2021-09-21 |
| Self-supervised Character-to-Character Distillation for Text Recognition | ✓ Link | 92.7 | CCD-ViT-Small | 2022-11-01 |
| Out of Length Text Recognition with Sub-String Matching | ✓ Link | 92.7 | SMTR (Union14M-L training) | 2024-07-17 |
| Focus on the Whole Character: Discriminative Character Modeling for Scene Text Recognition | ✓ Link | 91.8 | CFE-Base (synthetic training) (extra data) | 2024-07-08 |
| Self-supervised Character-to-Character Distillation for Text Recognition | ✓ Link | 91.6 | CCD-ViT-Tiny | 2022-11-01 |
| IterVM: Iterative Vision Modeling Module for Scene Text Recognition | ✓ Link | 90.9 | IterNet (3× IterVM, 3× IterLM) (extra data) | 2022-04-06 |
| Linguistic More: Taking a Further Step toward Efficient and Accurate Scene Text Recognition | ✓ Link | 90.9 | LPV-Base (extra data) | 2023-05-09 |
| Class-Aware Mask-Guided Feature Refinement for Scene Text Recognition | ✓ Link | 90.7 | CAM-Tiny | 2024-02-21 |
| Multi-modal Text Recognition Networks: Interactive Enhancements between Visual and Semantic Features | ✓ Link | 90.6 | MATRN | 2021-11-30 |
| Visual Semantics Allow for Textual Reasoning Better in Scene Text Recognition | ✓ Link | 90.6 | S-GTR | 2021-12-24 |
| Self-supervised Implicit Glyph Attention for Text Recognition | ✓ Link | 90.5 | SIGA_T | 2022-03-07 |
| Read Like Humans: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Recognition | ✓ Link | 90.1 | ABINet-LVst (extra data) | 2021-03-11 |
| CDistNet: Perceiving Multi-Domain Character Distance for Robust Text Recognition | ✓ Link | 89.77 | CDistNet (Ours) | 2021-11-22 |
| TextSSR: Diffusion-based Data Synthesis for Scene Text Recognition | ✓ Link | 89.61 | MAERec (ST + TextSSR-F 3.55M) (extra data) | 2024-12-02 |
| TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition | ✓ Link | 89.6 | ABINet-LV + TPS++ | 2023-05-09 |
| DiffusionSTR: Diffusion Model for Scene Text Recognition | | 89.2 | DiffusionSTR | 2023-06-29 |
| Look Back Again: Dual Parallel Attention Network for Accurate and Robust Scene Text Recognition | ✓ Link | 89.0 | DPAN | 2021-08-01 |
| On Recognizing Texts of Arbitrary Shapes with 2D Self-Attention | ✓ Link | 86.5 | SATRN (best checkpoint per benchmark) | 2019-10-10 |
| From Two to One: A New Scene Text Recognizer with Visual Language Modeling Network | ✓ Link | 86.0 | VisionLAN (extra data) | 2021-08-22 |
| MASTER: Multi-Aspect Non-local Network for Scene Text Recognition | ✓ Link | 84.5 | MASTER (h=8, N=3) (extra data) | 2019-10-07 |
| Vision Transformer for Fast and Efficient Scene Text Recognition | ✓ Link | 81.8 | ViTSTR-Base + RandAugment (extra data) | 2021-05-18 |
| SEED: Semantics Enhanced Encoder-Decoder Framework for Scene Text Recognition | ✓ Link | 81.4 | SE-ASTER (extra data) | 2020-05-22 |
| What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis | ✓ Link | 79.2 | TRBA (TPS-ResNet-BiLSTM-Attn) | 2019-04-03 |