Our IAMAS Lab and Aida Lab published a journal paper on using image processing techniques to generate pseudo-groundtruth to improve machine learning for classifying document images: Liu, Y. and L.-K. Soh (2025). Integrating Textual and Visual Features to Generate Pseudo-Groundtruth for Enhancing Training of Machine Learning Models in Document Image Classification, Journal of Electronic Imaging, 34(3), 033209, 23 May 2025, https://doi.org/10.1117/1.JEI.34.3.033029.