To content
Fakultät für Informatik

Two Papers Accepted at ICDAR 2025

Please provide a copyright notice
We’re thrilled to announce that two of our research papers have been accepted at the prestigious ICDAR 2025 conference in Wuhan, China.

We are delighted to share that two of our research papers have been accepted for presentation at the International Conference on Document Analysis and Recognition (ICDAR) 2025, taking place in Wuhan, China.

The first paper, "CM1 — A Dataset for Evaluating Few-Shot Information Extraction with Large Vision Language Models", is authored by Fabian Wolf, Oliver Tüselmann, Arthur Matei, Lukas Hennies, Christoph Rass, and Gernot A. Fink. It introduces a novel dataset designed to advance the evaluation of few-shot capabilities in modern Vision-Language Models, offering a valuable resource for future research in information extraction from documents.

The second paper, "Interpretable Writer Recognition via Vectors of Locally Aggregated Character", is the result of a collaboration between Tim Raven, Gernot A. Fink, and Vincent Christlein. This work proposes a transparent and effective method for writer recognition by leveraging character-level feature aggregation, contributing to the growing need for interpretability in document forensics and handwriting analysis.

Both papers represent significant steps forward in our mission to develop robust and explainable AI tools for document analysis.

We are honored by the recognition and are looking forward to engaging with the global research community at ICDAR 2025. See you in Wuhan!