What 70 Scientific Books Taught Me About Why AI Misreads a PDF

<p>My name is printed in the front matter of more than seventy scientific volumes, and I have spent more than twenty years turning manuscripts into typeset pages. I now build AI systems that read documents for a living. Those two facts sound unrelated. They are the same problem seen from opposite ends, and the first one explains most of what goes wrong in the second.</p> <p>Here is the short version. A document that has been typeset is not a bag of words. It is a structure, and almost all of that structure is thrown away the moment it becomes a PDF. When an AI reads that PDF, it is not reading the document. It is reading the shadow the document left on a page.</p> <h2> What a typesetter actually does </h2> <p>Setting a scientific book is not decoration. It is the act of encoding meaning into position.</p> <p>A running head tells you which chapter you are in without reading a word. A float, a table or figure that moves to wherever it fits, carries a caption precisely because it has been separated from the paragraph that referred to it. Front matter is a defined sequence: half title, title, copyright, dedication, contents, preface. A displayed equation is numbered because something later will point back at it. A footnote sits at the bottom of a page and a citation sits in a bibliography, and those are different objects doing different jobs.</p> <p>None of that is styling. Every one of those decisions is semantic. The compositor's craft is knowing that a two-line gap and a smaller point size mean <em>this is an aside, not the argument</em>, and that a reader will absorb that without ever being told.</p> <p>In LaTeX, the source knows all of this explicitly. There is a <code>\chapter</code>, a <code>\caption</code>, a <code>\label</code>, a <code>\cite</code>. The structure is not implied by how it looks. The structure is written down, and the appearance is derived from it.</p> <h2> What the PDF keeps </h2> <p>The PDF keeps the appearance and discards the source.</p> <p>A PDF is a set of instructions for placing glyphs at coordinates. Put this character at this x and y, in this font, at this size. That is essentially all it is. The fact that a particular run of glyphs was a chapter title, and another was a caption, and a third was a footnote, is not recorded anywhere. It is inferable by a human because humans read layout fluently. It is not stated.</p> <p>So when a system ingests that PDF, it has to reconstruct, by guesswork, everything the compositor originally knew for certain. Is this larger text a heading or an emphasised sentence? Does this two-column page read left column then right, or across? Is this number a footnote marker or an exponent? Does this table have a header row, or is the first row just bold?</p> <p>Every one of those guesses is a place where a document quietly becomes a different document.</p> <h2> The failures I see, and why they are structural </h2> <p>Three show up constantly, and none of them is a model problem.</p> <p><strong>Reading order.</strong> Two-column layouts are the classic. Extraction that follows the coordinate stream rather than the column flow will interleave the two columns line by line. The output is grammatical, fluent nonsense. Nothing errors. The text is all there. It is simply in an order that never existed.</p> <p><strong>Orphaned captions and floats.</strong> Because a float moves, a table can be printed pages away from the sentence that discusses it. Flatten that and the table arrives with no context, or worse, glued to whatever paragraph happened to be adjacent on the page. A figure that supports one claim gets silently attached to another.</p> <p><strong>Footnotes folded into the body.</strong> A footnote is a qualification, very often the exact qualification that limits the claim above it. Merge it into the running text and you get a sentence that asserts something the author deliberately hedged.</p> <p>I want to be precise about the point