bug/pdf-extraction-bug

[Orbiszeus]

Describe the bug
While partition_pdf or partition(text.. ) this method is working for docx, txt however for some pdfs it is not parsing well especially academic papers.

To Reproduce
elements = self._partition(**self._build_partition_kwargs(loaded))

Expected behavior
Should return a simple text inside the elements list with some metadata.

Screenshots

**Environment Info**
name = "unstructured"
version = "0.17.2"
description = "A library that prepares raw documents for downstream ML tasks."
optional = false
python-versions = ">=3.9.0"
groups = ["main"]
files = [
    {file = "unstructured-0.17.2-py3-none-any.whl", hash = ..
    {file = "unstructured-0.17.2.tar.gz", hash = ..

Additional context
I will upload the pdf example that is not working.

2025.findings-naacl.114.pdf