SemanticScuttle - klotz.me

klotz: spatial layout* + large language models* + fine-tuning* + docllm* + pre-training*

0 bookmark(s) - Sort by: Date ↓ / Title / - Bookmarks from other users for this tag

1. [2401.00908] DocLLM: A layout-aware generative language model for multimodal document understanding

   DocLLM is a lightweight extension to traditional LLMs for reasoning over visual documents, considering both textual semantics and spatial layout. It avoids expensive image encoders and focuses on bounding box information. It outperforms SotA LLMs on 14 out of 16 datasets across all tasks and generalizes well to previously unseen datasets.
   
   Keywords:

   2024-01-06 Tags: docllm, large language models, visual documents, textual semantics, spatial layout, pre-training, fine-tuning by klotz

Top of the page

First / Previous / Next / Last / Page 1 of 0