Document Analysis and Recognition - ICDAR 2025 Workshops
Description
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)
More details
Other editions
Additional editions
Content
.- The Fifth ICDAR International Workshop on Machine Learning (WML 2025)
.- PBa-LLM: Privacy- and Bias-aware NLP using Named-Entity Recognition (NER).
.- Benchmarking Graph Neural Networks for Document Layout Analysis in Public Affairs.
.- Improving Handwritten Text Recognition via 3D Attention and Multi-Scale Training.
.- Masked Self-Supervised Pre-Training for Text Recognition Transformers on Large-Scale Datasets.
.- Text Prompt to Image Generation for Classification of Similar and Non-Similar Scene Images to Improve Text Spotting Performance.
.- Enhancing Document VQA Models via Retrieval-Augmented Generation.
.- A New Multimodal Cross-Domain Network for Classifi-cation of Challenging Scene Images.
.- TextBite: A Historical Czech Document Dataset for Logical Page Segmentation.
.- Few-Part-Shot Font Generation.
.- Non-Linear Audio-Visual Storytelling from Scanned Comics: A Character-Centric Approach.
.- Automatic Text Box Placement for Supporting Typographic Design.
.- Visual Document Matching for Zero-Shot Document Classification.
.- Evaluating Popular Scene Text Detection and Recognition Methods on Tombstones.
.- Deep learning for defect detection in answer document image.
.- ResNet-TPP: A Parallel PHOC-PHOS Framework for Zero-Shot Handwritten Word Recognition in Low-Resource Scripts.
.- Interpret, prune and distill Donut :towards lightweight VLMs for VQA on documents.
.- Link prediction Graph Neural Networks for structure recognition of Handwritten Mathematical Expressions.
.- Rule-Based Reinforcement Learning for Document Image Classification with Vision Language Models.
.-
ICDAR 2025 Workshop on Multi-Modal Mathematical Reasoning in Documents (M3RD 2025).
.- Boosting Handwritten Mathematical Expression Recognition through Contextual Reasoning with Vision Large Language Models (vLLMs).
.- SCANS: An efficient Geometric Problem Solver with Content-Aware Attention and Adaptive Fusion.
.- GeoGRPO: Investigating the Stepwise-GRPO Enhancement in RLHF Framework.
.- Offline handwritten mathematical formula recognition based on primitive representation.
.- Long Math Reasoning Problem Generation.