pdf-translate

You can translate an academic PDF while preserving math and figures by using a pipeline that extracts text and images, retains LaTeX math, and renders HTML with MathJax and PDF with native Typst math.

PDF translation handles document structure and references by maintaining the original section, bibliography, and appendix hierarchy, while preserving English author names and citation references exactly as they appear in the source paper.

Yes, you can get both HTML and PDF outputs for translated academic papers. The process generates a standalone HTML for web deployment and a production-ready PDF for printing and publishing, both preserving the original document quality.

Translating PDFs with math and raster images does not reduce output quality, because the process preserves vector graphics and raster images at original quality while rendering LaTeX math natively through MathJax and Typst.

The best way to convert long-form technical papers for bilingual access is to use a robust pipeline that extracts text, preserves LaTeX math and figures, and generates ready-to-use HTML and PDF artifacts for researchers and publishers.

You do not need Typst or MathJax pre-installed to view the translated outputs. The process generates a standalone HTML that renders math via MathJax and a production-ready PDF that embeds native Typst math formulas directly.