eicker.news ᳇ tech newsOct 21
#Deepseek’s #OCR system compresses #image-based #textdocuments for language models, enabling AI to handle longer contexts. The system uses #DeepEncoder for #imageprocessing and a #textgenerator based on Deepseek3B-MoE, achieving up to tenfold compression while retaining 97% of the original information. https://the-decoder.com/deepseeks-ocr-system-compresses-image-based-text-so-ai-can-handle-much-longer-documents/?eicker.news #tech #media #news
Deepseek's OCR system compresses image-based text so AI can handle much longer documents

Chinese AI company Deepseek has built an OCR system that compresses image-based text documents for language models, aiming to let AI handle much longer contexts without running into memory limits.

THE DECODER