Document Analysis and Recognition – ICDAR 2025 Workshops: Wuhan, China, September 20–21, 2025, Proceedings, Part ILianwen Jin, Richard Zanibbi, Veronique Eglin The two-volume set LNCS 16225 + 16226 constitutes the proceedings of International Workshops co-located with the 19th International Conference on Document Analysis and Recognition, ICDAR 2025, held in Wuhan, China, during September 2025. The 46 full papers included in these proceedings were carefully reviewed and selected from a total of 74 submissions. The contributions stem from the following workshops: Part I: The Fifth ICDAR International Workshop on Machine Learning (WML 2025); ICDAR 2025 Workshop on Multi-Modal Mathematical Reasoning in Documents (M3RD 2025); Part II: The 16th IAPR International Workshop on Graphics Recognition (GREC 2025); ICDAR 2025 Workshop on Visual Text Generation and Text Image Processing(VT-TIP 2025); ICDAR 2025 Workshop on Documents Analysis of Low-resource Languages (DALL 2025) |
Contents
| 3 | |
| 21 | |
Improving Handwritten Text Recognition via 3D Attention and Multiscale Training | 36 |
Masked Selfsupervised Pretraining for Text Recognition Transformers on LargeScale Datasets | 53 |
Text Prompt to Image Generation for Classification of Similar and Nonsimilar Scene Images to Improve Text Spotting Performance | 71 |
Enhancing Document VQA Models via RetrievalAugmented Generation | 92 |
A New Multimodal CrossDomain Network for Classification of Challenging Scene Images | 108 |
A Historical Czech Document Dataset for Logical Page Segmentation | 124 |
Deep Learning for Defect Detection in Answer Document Image | 226 |
A Parallel PHOCPHOS Framework for ZeroShot Handwritten Word Recognition in LowResource Scripts | 245 |
Towards Lightweight VLMs for VQA on Documents | 260 |
Link Prediction Graph Neural Networks for Structure Recognition of Handwritten Mathematical Expressions | 279 |
RuleBased Reinforcement Learning for Document Image Classification with Vision Language Models | 292 |
ICDAR 2025 Workshop on Multimodal Mathematical Reasoning in Documents M3RD 2025 | 310 |
Boosting Handwritten Mathematical Expression Recognition Through Contextual Reasoning with Vision Large Language Models vLLMs | 313 |
An Efficient Geometric Problem Solver with ContentAware Attention and Adaptive Fusion | 327 |
FewPartShot Font Generation | 141 |
A CharacterCentric Approach | 158 |
Automatic Text Box Placement for Supporting Typographic Design | 175 |
Visual Document Matching for ZeroShot Document Classification | 192 |
Evaluating Popular Scene Text Detection and Recognition Methods on Tombstones | 209 |
Investigating the StepwiseGRPO Enhancement in RLHF Framework | 344 |
Offline Handwritten Mathematical Formula Recognition Based on Primitive Representation | 362 |
Long Math Reasoning Problem Generation | 378 |
| 394 | |
Other editions - View all
Document Analysis and Recognition – ICDAR 2025 Workshops: Wuhan, China ... Lianwen Jin,Richard Zanibbi,Veronique Eglin No preview available - 2025 |
Document Analysis and Recognition – ICDAR 2025 Workshops: Wuhan, China ... Lianwen Jin,Richard Zanibbi,Veronique Eglin No preview available - 2025 |
Common terms and phrases
accuracy algorithm Analysis and Recognition annotated approach architecture Artificial Intelligence arXiv preprint baseline benchmark bounding boxes challenges character classes classification Computer Vision Conference on Computer context convolutional CROHME dataset decoder defect detector Document Analysis document classification document image embeddings encoder enhance evaluation feature extraction feature map fine-tuning font framework fusion geometric graph GRPO GZSL handwritten mathematical expression IEEE improve International Conference labels large language models layers learning LLMs LNCS Machine Learning masked mathematical expression recognition methods metrics MixNet modalities module multimodal Named entity recognition neural network non-similar images optimal output patches Pattern Recognition performance pixels pre-training prediction Proceedings proposed Rand index reasoning reinforcement learning ResNet retrieval scene text segmentation self-attention self-supervised semantic sequence similar and non-similar spatial Speech Bubble structure Table tasks text lines text recognition text spotting textual tion tokens Transformer vector Vision and Pattern visual features VLMs Wang zero-shot Zhang