Poznámky k epizodě
MinerU-Diffusion reframes document OCR as inverse rendering, not language generation
This paper from Shanghai AI Lab and Peking University asks a simple systems question: if OCR is grounded in visual evidence, why should decoding still be forced into left-to-right token generation?
MinerU-Diffusion replaces autoregressive decoding with block-wise diffusion denoising under visual conditioning. The result is a better match to document OCR structure:
• up to 3.26x speedup over MinerU2.5
• 2.12x speedup at 99.9% relative accuracy
• 3.01x speedup at 98.8% relative accuracy
• stronger robustness…
Web: podcasters.spotify.com

