Recently we ran into an issue similar to what described in this post but with PDF documents. The SharePoint metadata “Title” is empty for these documents but the search results show some random text (which does not even present in the document) instead of document name. We have already implemented the fix for Word 2007 title issue described in the above mentioned post. On further investigating, it’s found that this behavior is due to the document’s internal property “Title” (not the document library metadata column “Title”). This title is set while generating the PDF files from a PDF creator/editor software. If SharePoint metadata “Title” is empty for the document, SharePoint crawler will consider the document’s internal property “Title” for the search result title. This can be verified by downloading the documents having incorrect title in search results and navigating to File > Properties after opening it in Adobe Reader. You can find the “Title” property is set to the text which is displayed in search results as title. So, I guess, the crawler considers text for search result title in the order of SharePoint metadata “Title”, document specific property “Title” and then file name if the first two are empty. I haven’t found a fix or workaround to change this order. The best option would be encouraging users to make good use of SharePoint metadata while sharing/collaborating on documents. Populating the SharePoint metadata associated with documents (especially “title”) will improve visibility of these documents in search.