@article{docs2synthasyntheticdatatrainedretriever, title = {Docs2Synth: A Synthetic Data Trained Retriever Framework for Scanned Visually Rich Documents Understanding}, author = {Yihao Ding and Qiang Sun and Puzhen Wu and Sirui Li and Siwen Luo and Wei Liu}, year = {2026}, eprint = {2601.12260}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2601.12260}, }