Parenthesis leads to duplicated pieces of text over the parsed file

### Question

Found a weird behavior of PDF parser. After process a PDF file (cyrillic text) I can see a lot of pieces of text everywhere.
Looks like the parser replaced every parenthesis "(" by the piece of text taken from title page (contains some copyright info).
I'm so sorry for image in russian but it shows the issue:

Finally, the orignal docoment is not so big, but the piece appears 159 times in the final text file (MD, JSON).
I'm not sure it is a bug or not. Of cource I can do some manual prapararations before processing the document. For example to remove the text from title page which is cause of the issue but I think it could be hard to process many documents.

Start by reproducing the report with a Cyrillic PDF in the PDF parser and inspect the generated Markdown and JSON for repeated title-page text after opening parentheses. Confirm whether the repetition occurs consistently, then trace the parser path responsible and verify that the duplicated text no longer appears in either output.

Written by the indexing model from the issue text.