Markdown does not capture the structure of a document - how can I fix this

[bukholm]

Markdown export does not reflect the hierarchic structure of the pdf source document i.e. all markdown headings have level one (##) . It is clear from the pdf document that there are chapters, sub chapters etc. both from text size, font, style and numbering.
So is there any way to tweak this or another model to use for identifying the document structure?

I'm using the code examples (export markdown) on the docling documentation page.

[corolair-hazem]

Facing same issue, Please let me know if you found a solution.

[PhanManhQuyet]

I have same issue. It does not capture document hierarchy.

[simjak]

same