L

Layout Xlm Base Finetuned With DocLayNet Base At Paragraphlevel Ml512

Developed by pierreguillou
This model is a fine-tuned version of the LayoutXLM base model on the DocLayNet dataset, specifically designed for document layout analysis and paragraph-level content understanding.
Downloads 79
Release Time : 3/25/2023

Model Overview

This is a multilingual document understanding model capable of identifying and analyzing paragraph-level elements in PDF documents, such as headings, text, tables, images, etc.

Model Features

Paragraph-Level Document Understanding
Capable of identifying and analyzing paragraph-level elements in documents, including 11 different types such as headings, text, tables, and images
Multilingual Support
Supports document analysis in multiple languages including English, German, French, and Japanese
High Accuracy
Achieved 86.55% paragraph accuracy and 96.93% token accuracy on the DocLayNet test set

Model Capabilities

Document Layout Analysis
Paragraph Classification
Multilingual Document Processing
PDF Content Understanding

Use Cases

Financial Document Processing
Financial Report Analysis
Automatically identifies different sections in financial reports, such as tables, text, and headings
Accuracy exceeds 90%
Legal Document Processing
Legal Clause Parsing
Identifies sections, clauses, and annotations in legal documents
Section heading recognition accuracy of 83.16%
Scientific Literature Processing
Scientific Paper Parsing
Identifies formulas and charts in papers
Formula recognition accuracy of 95.33%
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase