SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion

Re-assign community