Scripts and Data for "A Divergence-Driven Framework for Chinese Literary Text Digitization: Multi-Engine OCR Accuracy Evaluation and Multi-LLM Post-Correction"

This dataset contains Python scripts and experimental data supporting the paper "A Divergence-Driven Framework for Chinese Literary Text Digitization." It includes four experimental folders covering CER evaluation of OCR engines on Traditional and Simplified Chinese literary texts (Experiments 1 and 2), end-to-end four-stage workflow validation with multi-LLM arbitration (Experiment 3-A), and rule-based post-processing for Traditional Chinese text (Experiment 3-B).