dolphinbytedance
ADDEDByteDance's open-source document image parsing model (ACL 2025) that turns PDFs and photographed pages into structured Markdown/JSON, covering layout, reading order, text, tables, formulas, and code.
Discover quality open-source projects, submit projects anonymously, and claim and edit your own project.
A little curiosity. A world of open source.
THE FIRST COLLECTIONByteDance's open-source document image parsing model (ACL 2025) that turns PDFs and photographed pages into structured Markdown/JSON, covering layout, reading order, text, tables, formulas, and code.
Marked is a fast, lightweight Markdown parser and compiler that works in browsers, on Node.js servers, and via CLI. It supports all standard Markdown features, custom extensions, and is easily adaptable for different use cases.