Multimodal Ancient Document Parsing: Technical Report for EvaHan2026 Competition

Abstract

We present the multimodal Optical Character Recognition (OCR) and layout analysis methods developed for the EvaHan 2026 competition. Our approach is built upon the Qwen2.5-VL-7B-Instruct architecture and integrates two core strategies: (1) a reinforcement learning alignment pipeline utilizing Direct Preference Optimization (DPO) and Group Relative Policy Optimization (GRPO) to explicitly mitigate hallucination and coordinate instability; and (2) a four-stage curriculum learning framework that synthesizes domain-specific historical artifacts to enhance open-modality generalization. Using this approach, we achieve competitive results, notably reaching a Character Error Rate (CER) of 0.0303 on printed texts (Task A) and 0.0552 on handwritten manuscripts (Task C), as well as an Average Intersection over Union (IoU) of 0.7638 on layout element analysis (Task B).

Resources

Details