Mineru Diffusion Ocr Inverse Rendering

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.

adu2021 Updated 7 repo stars

File contents

adu2021/skillxiv commit d4e852c9a4

Frequently asked questions

npx get-skill add adu2021/mineru-diffusion-ocr-inverse-rendering