At the International Conference on Document Analysis and Recognition (ICDAR) 2025, two papers co-authored by Professor Wei Hongxi and his Ph.D. student Wang Yiming from the School of Artificial Intelligence, Inner Mongolia University, received prestigious awards, marking a significant achievement for the university in the field of document analysis and recognition.
Held from September 16 to 21, 2025, in Wuhan, ICDAR is the flagship conference of the International Association for Pattern Recognition (IAPR) in the field of document analysis. This year, the conference received 314 valid submissions, with 40 papers accepted for oral presentations (acceptance rate: 12.7%) and 102 for poster presentations (acceptance rate: 32.5%).
The two awarded papers are:
“DCC: Plug-and-Play Dynamic Category Compression for Enhanced Handwritten Text Generation”
🏆 Best Student Paper Runner-up Award
“VMF-Net: Visual-Aware Multi-representation Fusion Network for Artifact-Free Handwritten Mathematical Expressions Generation”
🏆 Best Poster Runner-up Award
Both works were led by Wang Yiming, a first-year Ph.D. student (Class of 2023), under the supervision of Prof. Wei Hongxi.
A Legacy of Excellence in Document Analysis
The School of Artificial Intelligence has a long-standing tradition of research excellence in document analysis and recognition, particularly in Mongolian script processing. Since the 1990s, the university’s Mongolian Intelligent Information Processing Team has been a pioneer in the field, developing:
Mongolian printed text OCR systems,
Mongolian online handwriting OCR,
OCR systems for ancient Mongolian manuscripts, among others.
Recognition and Impact
These award-winning papers not only enhance the international academic visibility of Inner Mongolia University in the domain of document recognition, but also reflect the school’s strong performance in postgraduate training and innovation in AI-driven text generation and mathematical expression recognition.
