arrow_back Guide
The Editor
What belongs on each tab, and the one rule that ties them together.
Read this first: why there are tabs at all
In Word you type everything into one long page, including the figures.
TeXpress splits your manuscript by what each thing is instead.
Your prose is on the Content tab. Your figures are on the Figures tab. Your equations are on the Equations tab.
Nothing is lost and nothing is hidden, it is just filed by type.
This is what lets one journal switch reformat the whole paper. A figure that knows it is a figure can be
given a number, a caption position and a column span by whichever journal you pick.
A picture pasted into a paragraph cannot.
The cost of that is one new habit: you say where a figure belongs by dropping a marker into your text,
rather than by dragging the image there. Those are the Fig, Tab and Eq buttons, and they are the next section.
Markers: Fig, Tab, Eq, Cite and Ref
Put your cursor in a paragraph on the Content tab, press one of these buttons in the toolbar,
pick an item from the list that opens ("Insert a citation", "Insert a figure reference" and so on), and a small coloured chip appears at the cursor.
The chip is a single object: you can delete it, but you cannot type inside it.
Fig Tab Eq
The Place buttons.
Each refers to a figure, table or equation. In the PDF it prints as "Fig. 2" or "Table 1",
numbered by the journal's own rules, and it tells LaTeX that this is where the item belongs.
The first marker for an item is also what puts it into the document; later ones are just cross-references to it.
Cite
Cites a reference from your bibliography. Prints in the journal's citation style,
[3] or (Smith, 2020) depending on the journal, and adds that reference
to the bibliography at the end. Citing is what decides whether a reference is printed at all.
Ref
Refers back to a theorem, lemma, definition or algorithm you inserted earlier,
even one in a different section. Prints as its number.
The rule that catches everyone once:
a figure or table you never place is added at the very end of the body rather than dropped, which looks wrong but is recoverable.
An equation you never place with the Eq button is not rendered at all.
An equation is not a float, so there is no sensible place to park it. If an equation is missing from your PDF, this is why.
segment Content
Your actual paper. Where you will spend most of your time.
Sections
- Each section is a card: a heading you can click and rename, and a body you type into. Click the header row to fold it away,
and drag a card to reorder it.
- The button on the right of a card adds a child section under it.
The indent and outdent arrows move an existing section a level in or out, and the up and down arrows reorder it.
Nesting can go up to six levels, or fewer if your journal allows fewer, and the button simply disappears at the limit.
- The grey number to the left of a heading ("2.1") is an outline aid so you can see the shape of your paper.
The real numbering in the PDF comes from the journal.
- To make a section an appendix, click into it, open the ⋮ menu at the right of the toolbar and tick
Move to appendix. It moves to the back matter with the journal's appendix numbering.
- A section with a single subsection gets a short amber note at the top of the tab. Journals usually query that at proof stage,
so it is worth adding a second subsection, or folding the lone one into its parent, before you submit.
- Do not create a References section here.
One is built automatically from the References tab. A hand-made one would be a second, duplicate list.
The toolbar, and writing in the body
- One toolbar serves the whole tab and stays pinned at the top while you scroll. It acts on whichever section your cursor is in,
and names that section at its right-hand end ("Click a section to edit" until you have).
- It gives bold, italic, superscript, subscript, bullet lists and numbered lists, then the marker buttons
(Fig, Tab, Eq, Cite, Ref), Env
and Math. What you see is what you get, there is no markup to learn.
- TeX opens the active section as LaTeX (see the end of this page).
- Find & replace is in the ⋮ menu, or press Ctrl+F.
It steps through matches with a counter and a highlight, and replaces one or all. It only ever changes your prose:
it never touches the inside of a citation, a cross-reference or inline maths, so a replace cannot quietly break your bibliography.
- Math inserts an inline symbol or expression using the visual
equation builder. Use it for a lone λ or x² in a sentence.
Full displayed equations belong on the Equations tab.
- Pasting from Word or a PDF is cleaned up: formatting we support is kept, everything else is dropped rather than smuggled in.
Environments (theorems, proofs, algorithms)
Env in the toolbar opens a two-step window:
pick the type, then fill it in. The available types are Theorem, Lemma, Corollary, Definition, Proposition,
Proof, Remark, Example, Conjecture and Algorithm.
- Every environment needs a label, which is the short name you use
to refer back to it (
thm:main-result). Letters, digits, underscore, hyphen and colon, no spaces.
An optional title prints in brackets after the number.
- Numbering ("Theorem 1", "Lemma 2") is applied at compile time by the journal. Do not type numbers yourself.
- The environment is anchored where your cursor was, and you can refer back to it anywhere with the
Ref button.
- Algorithms are entered line by line, not as a block of text.
Each pseudocode step is its own line with its own indent control, so "for each i" and the two steps inside it are three lines at two
indent levels. That structure is what lets us emit real numbered pseudocode in whichever algorithm package your journal uses,
instead of a screenshot-shaped code block. Insert equation puts inline
math into the line you are on.
Two things that apply across every tab
Labels are unique across the whole project
A figure, a table, an equation and an environment cannot share a label, even though they are different kinds of thing.
If you try, you are told which item already owns that name. This is what makes a cross-reference unambiguous,
and it is why a habit of prefixes (fig:, tab:,
eq:, thm:) is worth adopting.
Tab as LaTeX, for people who want it
The Tab as LaTeX switch at the right of the tab row shows the current tab as plain, journal-neutral LaTeX. On Content, Tables, Equations,
Metadata, Declarations and References you can also edit that LaTeX and press
Apply to visual editor to turn it back into rows,
which makes pasting a chunk of an old paper very quick. Figures are read-only there, since no amount of text can restore an image.
The TeX button does the same for one item: on the Content toolbar
for the section you are in, and on each table and declaration card.
Two safety rules you cannot switch off: an apply that would empty a populated tab is refused rather than obeyed
(a parse that found nothing is a failed parse, not a delete instruction), and every apply saves a snapshot first,
so a bad paste is always recoverable from Activity. This is not the same thing as the manual LaTeX editor you reach through
Generate, which is covered on the next page.
Working with co-authors
Share in the top bar invites someone by email as either an
editor or a viewer, and a "Shared with" pill beside the project name shows how many people have access. Anyone in the project sees a banner when another person is working in it at the same time.
A shared project stays on the owner's plan: compiles and AI actions spend the owner's quota, never the collaborator's,
and a viewer can read every tab and download the compiled PDF without being able to change anything.
Still stuck?
The FAQs cover plans, quotas and data questions.
For anything else, tell us what happened and we will look at your project.
mail Contact us
