mineru-diffusion-ocr-inverse-rendering

【文档与办公】Replace autoregressive token-by-token OCR decoding with block-wise diffusion decoding to achieve 3.2× speedup while maintaining 99.9% accuracy. Works best for document layout parsing where parallel generation is feasible. Trigger: When optimizing OCR systems and want faster inference without accuracy loss.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准