Fetching the paper…

DocLLM: A layout-aware generative language model for multimodal document understanding · Around