generator-latex-template 2026.9.28 → 2026.10.8

package/README.md

@@ -132,7 +132,8 @@ In the long run, the contents of the `paper.tex` (and similar) files in reposito
 - Cancel workflows:
   - `gh run list -L 100 --json databaseId -s queued -R latextemplates/generator-latex-template | jq -r '.[] | .databaseId' | awk '{gsub(/\\r\\n/,RS)} 1' | while read -r run_id; do gh run cancel "$run_id" -R latextemplates/generator-latex-template || true; done`
   - `gh run list -L 100 --json databaseId -s in_progress -R latextemplates/generator-latex-template | jq -r '.[] | .databaseId' | awk '{gsub(/\\r\\n/,RS)} 1' | while read -r run_id; do gh run cancel "$run_id" -R latextemplates/generator-latex-template || true; done`
-- Update submodule of "derived" templates (LNCS, ...): `scripts/spread-updates.sh`
+- Update submodule of "derived" templates (LNCS, ...): `scripts/spread-updates.sh`. It also links the generator's `refine-ltg` PR in the description of each template's "Update LTG" PR.
+- Keep the `refine-ltg` PR description current: it links every issue the cycle fixes (the same `[repo#N]` links as the CHANGELOG).
 - Sort `dot.aspell.en.pws`: `cat dot.aspell.en.pws | (sed -u 1q; sort) | sponge dot.aspell.en.pws` <!-- hint by https://stackoverflow.com/a/56151840/873282 -->
 
 ### Testing
@@ -231,7 +232,7 @@ ejslint.cmd c:\git-repositories\latextemplates\generator-latex-template\generato
    Use [release-it](https://www.npmjs.com/package/release-it) (do not create a release on GitHub) and [github-release-from-changelog](https://www.npmjs.com/package/github-release-from-changelog).
 
    - `npx release-it`
-   - `npx github-release-from-changelog`
+   - `GITHUB_TOKEN=$(gh auth token) npx github-release-from-changelog` (Git Bash; it needs a GitHub token, which the logged-in `gh` CLI provides)
 5. On `main`, run `scripts/end-new-cycle.sh`, then **squash-merge** each template's "Update LTG" PR once it is green.
 
 ## License

package/generators/app/CLAUDE.md

@@ -0,0 +1,113 @@
+<!-- Loaded automatically when Claude works on files under generators/app/. Moved from the repository root CLAUDE.md. -->
+
+## Adding a package or example (wire every place, or a variant silently drops it)
+
+The generated document is assembled from micro-templates; a new LaTeX package or
+example section has to be wired in **lockstep** across several files. Miss one and
+the feature vanishes for some switch combination without any error.
+
+1. **Content** — add/extend `generators/app/templates/<name>.example.<lang>.tex`
+   (the demo body) and/or `<name>.preamble.<lang>.tex` (the `\usepackage`). Example
+   bodies use the `<%- bexample %> … <%- eexample %>` convention (=
+   `\begin{ltgexample}…\end{ltgexample}`), which **runs the body in place and echoes
+   its own source lines** — it is *not* the old `latexdemo`/`\PrintDemo`. Most examples
+   are a single shared `.en` file included from both mains. Give each feature its **own**
+   `<feature>.preamble.<lang>.tex` (one concern per file — e.g. `tikz.preamble`,
+   `longtable.preamble`, `uml.plantuml.preamble`), **not** a catch-all "examples" preamble.
+   **Config vs. demo:** a package a feature needs to *work* belongs in that feature's
+   config preamble and is loaded wherever the feature is on; a package that only *renders
+   a demo* is the example's own concern. So `\mathbb` (a real feature) is provided by
+   `math.preamble.en.tex` — `\@ifundefined{mathbb}{\usepackage{amssymb}}{}`, guarded
+   because `amssymb` clashes with `mathdesign`/`newtxmath`/`unicode-math` — and the math
+   *example* is then gated to where `math.preamble` is included (`isThesis ||
+   feature.abbreviations`), so LNCS etc. don't reference an undefined `\mathbb`.
+2. **Wire the includes into `main.en.tex` AND `main.de.tex`.** Example includes go in
+   the `LaTeX Hints` chapter (the whole chapter is already guarded by
+   `<% if (examples) %>`); package loads go in the preamble area. Guard each with the
+   right condition (`isThesis`, a switch value, …) — and gate an example to where the
+   config that makes it compile is present.
+3. **`Texlivefile`** — add the TeX Live package name, guarded (usually
+   `githubpublish || <condition>`; `githubpublish` deliberately bundles everything).
+4. **Shell-escape is derived, never hardcoded.** It is the boolean
+   `requiresShellEscape` in `index.js` (`listings == "minted" || uml == "plantuml"`).
+   Extend that expression if the new package needs `\write18`; the `latexmkrc` and the
+   `main.*.tex` editor hints read the flag.
+5. **New user switch?** Decide **prompt vs. plain flag** first:
+   - A **prompt** in `options.js` (scope with `when()`, give a `default`) is interactive
+     and nice for discovery — but **every** template's `update-files.yml` that satisfies
+     its `when()` must pass the flag (like `--todo`), or the non-interactive `yo` run in CI
+     blocks on the prompt and fails. So a thesis-only prompt (e.g. `uml`) means
+     `scientific-thesis-template` *and* `uni-stuttgart-dissertation-template` need
+     `--uml=…` added to their (template-managed) `update-files.yml`.
+   - A **plain CLI flag** read from `this.options` in `index.js` with a default (e.g.
+     `longtable`) never prompts, so it never breaks CI and needs no `update-files.yml`
+     change — at the cost of interactive discovery.
+   Either way default it in `index.js` (prompt: `if (!this.props.x) this.props.x = "none";`
+   / flag: read `this.options.x` with a fallback), and if it's a prompt add the axis to
+   `__tests__/matrix.js` **and** the list in `.github/generate-workflows.py` (kept in
+   sync). Model on the `todo` / `uml` (prompt) and `longtable` (flag) options.
+6. **Not in TeX Live** (e.g. `tikz-uml`) — guard usage with
+   `\IfFileExists{<pkg>.sty}{…}{…}` so a plain `npx`/submodule-less generation still
+   compiles (it just skips the demo). Provide the package via a git submodule +
+   `TEXINPUTS` (in `latexmkrc`) and a recursive checkout in the generated `check.yml`,
+   and document the setup in `README.en.md`. External-tool packages (PlantUML needs
+   Java) also need the tool installed in `check.yml` and `--shell-escape`.
+
+**Verify (LaTeX-free first, then compile):** `npx ejs-lint <file>` on every changed
+EJS template; run **`npm test`**, not `node --test` directly — `npm test`'s `pretest`
+runs `eslint .` (incl. `prettier`), which `node --test` skips, so a prettier-only nit
+in `index.js`/`options.js` passes locally but fails CI (run `npx eslint --fix .` to
+match the formatter). `npm test` is the pairwise suite; `npm run test:all` is the full
+matrix. Then generate locally with the **exact** `update-files.yml` flags (a missing
+required option such as `acmformat` or `todo` drops you into an interactive prompt,
+which reads as a hang) and grep the output. Sanity-compile new `.tex` in a *minimal
+standalone* document (the full thesis pulls in packages a local TeX Live may lack);
+`perl -c` the generated `latexmkrc`; validate the generated `check.yml` as YAML. The
+`check-make.yml` build also runs `make format` (latexindent) + `git diff --exit-code`,
+so generated output must already be latexindent-clean — column-aligned tables in
+particular (see `lookForAlignDelims` in `localSettings.yaml`).
+
+### EJS whitespace: `-%>` eats the trailing newline
+
+Almost every control tag ends with `-%>` (the EJS "newline slurp"): `<% if (…) { -%>`,
+`<% } -%>`, `<% switch/case/break … -%>`, and `<%- include('…', this); -%>`. That slurp
+is deliberate — it keeps the control flow (and the include statements themselves) from
+emitting stray blank lines into the generated `.tex`. A plain `<% } %>` (no dash) leaves
+its newline and can inject an unwanted blank line, so the convention is the dashed form;
+only use plain `%>` when you actually want that newline.
+
+The consequence (and a real footgun): **the blank-line layout of the output is not the
+layout of the source.** Because `-%>` eats one trailing newline, two adjacent
+`-%>`-terminated blocks/includes render with **no** blank line between them. To get a
+blank line between blocks you need one of:
+- the first block's file to **end with a trailing blank line** (how most
+  `*.preamble.*.tex` separate themselves — e.g. `font.preamble` ends with a blank line
+  then `<% } -%>`), **or**
+- an **explicit blank line in the source** between the two `-%>` tags (the `-%>` eats one
+  newline, the remaining one + the content's own trailing newline = a blank line).
+
+This is exactly the `nowidow` bug: it followed `microtype.preamble`, which does *not* end
+with a trailing blank, and the include `-%>` ate the source newline — so the block butted
+against microtype until an explicit blank line was added between the two includes. When a
+generated block is missing/extra a blank line, this slurp interaction is the first thing
+to check.
+
+### Compile gotchas the single-variant local test misses
+
+`npm test` only checks that the generator *generates*; the **multi-variant template
+builds** (each `update-ltg` PR compiles every `main*.tex` for both languages and several
+formats) catch things a single local `yo` + compile does not. Two that have bitten:
+
+- **Two-column layouts can't use `longtable`** ("longtable not in 1-column mode"). IEEE and
+  the ACM `sigconf`/`sigplan`/`acmtog` formats are two-column, so any `longtable` (or other
+  one-column-only) example must be gated on `!twocolumn` (a derived prop in `index.js`:
+  `documentclass == "ieee" || (acmart && acmformat ∈ {acmtog,sigconf,sigplan})`). Single-column
+  figures (`figure`, not `figure*`) are fine in two columns.
+- **A template repo ships ONE shared `abbreviations.tex`** — the `update-files.yml` "Prepare
+  files" step keeps only one variant's copy (e.g. the `en`+minted one), and **every** main
+  file (English *and* German) does `\input{abbreviations}`. So a glossary entry used by any
+  language's example must be defined in **both** `abbreviations.en.tex` and
+  `abbreviations.de.tex` (the unused one just never appears in that document's list). Making
+  `\gls` entries language-exclusive breaks the other language's variant with
+  "Glossary entry `…' has not been defined". The same single-shared-file logic applies to
+  other `this.fs.copy`'d siblings (`commands.tex`, …).

package/generators/app/templates/README.en.md

@@ -670,6 +670,13 @@ The author of the class offers a large FAQ at <http://www.michaelshell.org/tex/i
 Please read on there.
 The other possibility is to execute `texdoc ieeetran` and read in the documentation.
 For example, there is an explanation of how to typeset the affiliation information with four or more authors properly.
+
+### Q: How are the two columns of the last page balanced?
+
+The [flushend](https://ctan.org/pkg/flushend) package balances them automatically; no command in the text is needed.
+An `\AtEndDocument` hook in the preamble (right after `\usepackage{flushend}`) activates flushend for the last page and, when the text ends in the first column, ends that column, so that flushend distributes the text over both columns.
+If you do not want balanced columns, remove `\usepackage{flushend}` together with that hook.
+If flushend produces odd spacing on the last page (IEEEtran's author warns about that for references), balance by hand instead: remove flushend and enable `\IEEEtriggeratref{N}` in front of `\bibliographystyle`, which starts a new column before reference `N`.
 <% break; case "ustutt": -%>
 
 ### Q: How to rename `thesis-example.tex`?

package/generators/app/templates/Texlivefile

@@ -235,6 +235,9 @@ nowidow
 ntheorem
 <% } -%>
 paralist
+<% if (githubpublish || twocolumn) { -%>
+pdfcol<%# required by the tcolorbox library "breakable" (example boxes in two-column documents) %>
+<% } -%>
 pdfcomment
 pdfescape
 <% if (githubpublish || isThesis) { -%>
@@ -253,8 +256,8 @@ placeins
 <% if (githubpublish || (uml == "plantuml")) { -%>
 plantuml<%# plantuml.sty for the PlantUML UML example; the diagram also needs an external PlantUML install %>
 <% } -%>
-<% if ((documentclass == 'acmart') || (documentclass == 'ieee')) { -%>
-preprint<%# this includes balance%>
+<% if (documentclass == 'acmart') { -%>
+preprint<%# this includes balance, used by the acmart option "balance"%>
 <% } -%>
 <% if (githubpublish || isThesis) { -%>
 psnfss<%# provides pifont%>
@@ -290,7 +293,7 @@ soulpos
 spelling
 <% } -%>
 stringenc
-sttools<%# this includes stfloats; at least required by acmart, ieee, lncs, ustutt, scientific-thesis %>
+sttools<%# this includes stfloats (at least required by acmart, ieee, lncs, ustutt, scientific-thesis) and flushend (ieee: balances the columns on the last page) %>
 subfig
 svn-prov<%# at least required by acmart, ustutt, scientific-thesis%>
 <% if (githubpublish || isThesis) { -%>

package/generators/app/templates/diagbox.example.en.tex

@@ -1,14 +1,15 @@
+<%# Keep the first cell of the table short: latexindent pads the first cell of every row to its width, and the echoed code has to fit a narrow column (IEEE, ACM). -%>
 <%- bexample %>
-% Source: https://tex.stackexchange.com/a/468994/9075
+% Adapted from https://tex.stackexchange.com/a/468994/9075
 \begin{<% if (documentclass == 'ieee') { %>figure<% } else { %>table<% } %>}
   \caption{Table with diagonal line}
   \label{tab:diag}
   \begin{center}
     \begin{tabular}{|l|c|c|}
       \hline
-      \diagbox[width=10em]{Diag \\Column Head I}{Diag Column\\Head II} & Second & Third \\
+      \diagbox{Row}{Column} & A & B \\
       \hline
-                                                                       & foo    & bar   \\
+      1                     & x & y \\
       \hline
     \end{tabular}
   \end{center}

package/generators/app/templates/dot.aspell.conf

@@ -1,5 +1,6 @@
 add-tex-command "minted options" ppppppppppppppppppppppp
 add-tex-command @ifundefined pppppp
+add-tex-command AtEndDocument p
 add-tex-command Cref p
 add-tex-command DeclareFontFamily ppp
 add-tex-command DeclareFontShape pppppp

package/generators/app/templates/dot.textlintrc.json

@@ -4,7 +4,7 @@
   },
   "filters": {
     "allowlist": {
-      "allow": ["/\\bonly\\b/i"]
+      "allow": ["/\\bonly\\b/i", "/\\bFinally,/"]
     }
   },
   "rules": {

package/generators/app/templates/hyphenation.example.en.tex

@@ -7,16 +7,15 @@ Here are some hints:
 
 <%- bexample %>
 In case you write <%- bquote %>application-specific<%- equote %>, then the word will only be hyphenated at the dash.
-You can also write \verb1applica\allowbreak{}tion-specific1 (result: applica\allowbreak{}tion-specific), but this is much more effort.
+You can also insert \verb1\allowbreak{}1 into the word, as in applica\allowbreak{}tion-specific, but this is much more effort.
 
 <% if (tweakouterquote == "outerquote")  { -%>
 In the net, you will read about that one can use a normal quote sign followed by an equal sign (\verb1"=1) to specify that hyphenation is allowed at other places in the word.
 Since we use the quote sign for automatic quotation, this is not possible anymore.
-Please use the command \verb1\hydash1 for the same purpose.
-Example: \verb1application\hydash specific1 (result: application\hydash specific).
+Please use the command \verb1\hydash1 for the same purpose, as in application\hydash specific.
 <% } else { -%>
 You can now write words containing hyphens which are hyphenated at other places in the word.
-For instance, \verb1application"=specific1 gets application"=specific.
+For that, write \verb1"=1 instead of \verb1-1, as in application"=specific.
 This is enabled by an additional configuration of the babel package.
 <% } -%>
 <%- eexample %>

package/generators/app/templates/latexmkrc

@@ -7,14 +7,34 @@ $pdflatex = '<%= reallatexcompiler %> <% if (requiresShellEscape) { %>--shell-es
 $ENV{'TEXINPUTS'} = './tikz-uml//:' . ($ENV{'TEXINPUTS'} // '');
 <% } -%>
 
-#Use SumatraPDF instead of the default PDF viewer
-$pdf_previewer = 'start "C:\Program Files\SumatraPDF\SumatraPDF.exe"';
+#------------------------------------------------------------------------------#
+# compilation
 
-#SumatraPDF updates automatically
+# automatically call lualatex/pdflatex (instead of latex)
+$pdf_mode = 1;
+
+# replaces main.pdf by jobname.pdf
+# Note: -pvc may then not find the PDF.
+#$jobname = 'thesis';
+
+#------------------------------------------------------------------------------#
+# preview
+# You may put these settings in ~/.latexmkrc instead.
+
+# >0 <-> equivalent to -pv
 $preview_mode = 0;
+# >0 <-> equivalent to -pvc
+#$preview_continuous_mode = 1;
 
-#automatically call lualatex/pdflatex (instead of latex)
-$pdf_mode = 1;
+if ($^O eq 'MSWin32') {
+  # Use SumatraPDF instead of the default PDF viewer; it reloads the PDF automatically
+  $pdf_previewer = 'start "C:\Program Files\SumatraPDF\SumatraPDF.exe"';
+}
+# 'evince' is 'Document Viewer' (GNOME's default)
+#$pdf_previewer = 'evince';
+
+#------------------------------------------------------------------------------#
+# bibliography and glossaries
 
 # enable deletion of *.bbl when calling "latexmk -c"
 $bibtex_use = 2;
@@ -33,6 +53,9 @@ sub run_makeglossaries {
   };
 }
 
-#remove more files than in the default configuration
+#------------------------------------------------------------------------------#
+# cleanup
+
+# remove more files than in the default configuration
 @generated_exts = qw(acn acr alg aux code fls glg glo glo-abr gls glsdefs idx ind ist lof lol lot out run.xml synctex thm toc tpt upa upb wrt);
 $clean_ext .= ' %R.ist %R.xdy';

package/generators/app/templates/listings.example.de.tex

@@ -62,3 +62,7 @@ public class Hello {
 }
 \end{lstlisting}
 <%- eexample %>
+<% if (isThesis) { -%>
+
+<%- include('listings.long.example.en.tex', this); -%>
+<% } -%>

package/generators/app/templates/listings.example.en.tex

@@ -62,3 +62,7 @@ public class Hello {
 }
 \end{lstlisting}
 <%- eexample %>
+<% if (isThesis) { -%>
+
+<%- include('listings.long.example.en.tex', this); -%>
+<% } -%>

package/generators/app/templates/listings.long.example.en.tex

@@ -0,0 +1,75 @@
+Listings longer than the remaining space continue on the next page if they are \emph{not} floating.
+A floating listing is always placed on a single page.
+<%- Cref %>{lst:long} spans multiple pages.
+
+\begin{lstlisting}[
+  language=Java,
+  caption={Long Java listing spanning multiple pages},
+  label={lst:long}]
+public class Squares {
+
+    /** Returns the square of 1. */
+    public int square1() {
+        return 1 * 1;
+    }
+
+    /** Returns the square of 2. */
+    public int square2() {
+        return 2 * 2;
+    }
+
+    /** Returns the square of 3. */
+    public int square3() {
+        return 3 * 3;
+    }
+
+    /** Returns the square of 4. */
+    public int square4() {
+        return 4 * 4;
+    }
+
+    /** Returns the square of 5. */
+    public int square5() {
+        return 5 * 5;
+    }
+
+    /** Returns the square of 6. */
+    public int square6() {
+        return 6 * 6;
+    }
+
+    /** Returns the square of 7. */
+    public int square7() {
+        return 7 * 7;
+    }
+
+    /** Returns the square of 8. */
+    public int square8() {
+        return 8 * 8;
+    }
+
+    /** Returns the square of 9. */
+    public int square9() {
+        return 9 * 9;
+    }
+
+    /** Returns the square of 10. */
+    public int square10() {
+        return 10 * 10;
+    }
+
+    /** Returns the square of 11. */
+    public int square11() {
+        return 11 * 11;
+    }
+
+    /** Returns the square of 12. */
+    public int square12() {
+        return 12 * 12;
+    }
+
+    public static void main(String[] args) {
+        System.out.println(new Squares().square12());
+    }
+}
+\end{lstlisting}

package/generators/app/templates/longtable.example.en.tex

@@ -1,7 +1,7 @@
 <%= heading2 %>{Tables spanning multiple pages}
 
 The \href{https://ctan.org/pkg/longtable}{longtable} package typesets tables that automatically break across page boundaries.
-The header (\verb+\endhead+) and footer (\verb+\endfoot+) rows are repeated on every page; once enough rows are added, the table spans several pages on its own.
+The header (\verb+\endhead+) and footer (\verb+\endfoot+) rows are repeated on every page; once enough rows are added, the table spans more than one page on its own.
 
 <%- bexample %>
 \begin{longtable}{|l|l|l|}

package/generators/app/templates/main.de.tex

@@ -56,9 +56,18 @@
 % For the final version, replace "draftcls" by "final"
 \documentclass[<%= ieeevariant %>,<%= papersize %>paper,ngerman]{IEEEtran}[2015/08/26]
 
-% Balance the last page using the balance package (see https://ctan.org/pkg/balance)
-% Alternative to balance is the pbalance package (see https://ctan.org/pkg/pbalance), which sometimes works better
-\usepackage{balance}
+% Balance the columns on the last page using the flushend package (see https://ctan.org/pkg/flushend).
+% It needs no command in the text.
+% The balance package (see https://ctan.org/pkg/balance) needs \balance in the first column of the last page, otherwise it warns "You have called \balance in second column".
+% The pbalance package (see https://ctan.org/pkg/pbalance) does not work together with stfloats, which is loaded for \fnbelowfloat.
+% IEEEtran's own way is \IEEEtriggeratref, see the bibliography section at the end of this file.
+\usepackage{flushend}
+% flushend activates its balancing output routine only right before the final \clearpage, which is too late in some documents (e.g., the German example paper under LuaLaTeX): the last page is shipped out before. Thus, \flushcolsend activates it already here.
+% flushend only balances when the text of the last page reaches the second column.
+% If the text ends in the first column, end that column and put an empty box into the second one, so that flushend distributes the text over both columns.
+\makeatletter
+\AtEndDocument{\flushcolsend\if@firstcolumn\ifdim\pagetotal>\z@\newpage\hbox to\columnwidth{}\fi\fi}
+\makeatother
 
 <% break; case "scientific-thesis": -%>
 
@@ -291,7 +300,7 @@
 <% if (useExampleEnvironment) { -%>
 \usepackage[realmainfile]{currfile}
 \usepackage{tcolorbox}
-\tcbuselibrary{<%= listings %>}
+\tcbuselibrary{<%= listings %><% if (twocolumn) { %>,breakable<% } %>}
 <% } -%>
 <% } -%>
 
@@ -376,7 +385,7 @@
 <% if (thesisvariant == "university-of-hamburg") { -%>
 \usepackage{geometry}
 <% } -%>
-<%- include('scientific-thesis-cover.preamble.en.tex', this); %>
+<%- include('scientific-thesis-cover.preamble.de.tex', this); %>
 <% break; } -%>
 <% if ((latexcompiler == "pdflatex") || (latexcompiler == "both")) { indent = '' -%>
 
@@ -609,7 +618,7 @@ Kurzfassung der Arbeit.
   In one sentence, what’s the topic?
   Phrase it in a way that your reader will understand.
   If you’re writing a PhD thesis, your readers are the examiners – assume they are familiar with the general field of research, so you need to tell them specifically what topic your thesis addresses.
-  Same advice works for scientific papers – the readers are the peer reviewers, and eventually others in your field interested in your research, so again they know the background work, but want to know specifically what topic your paper covers.
+  Same advice works for scientific papers – the readers are the peer reviewers, and later others in your field interested in your research, so again they know the background work, but want to know specifically what topic your paper covers.
 
   \emph{State the problem you tackle.}
   What’s the key research question?
@@ -620,12 +629,12 @@ Kurzfassung der Arbeit.
   Keep working at this step until you have a single, concise (and understandable) question.
 
   \emph{Summarize (in one sentence) why nobody else has adequately answered the research question yet.}
-  For a PhD thesis, you’ll have an entire chapter, covering what’s been done previously in the literature.
+  For a PhD thesis, you’ll have an entire chapter, covering the prior work in the literature.
   Here you have to boil that down to one sentence.
-  But remember, the trick is not to try and cover all the various ways in which people have tried and failed; the trick is to explain that there’s this one particular approach that nobody else tried yet (hint: it’s the thing that your research does).
+  But remember, the trick is not to try and cover all the ways in which people have tried and failed; the trick is to explain that there’s this one particular approach that nobody else tried yet (hint: it’s the thing that your research does).
   But here you’re phrasing it in such a way that it’s clear it’s a gap in the literature.
   So use a phrase such as “previous work has failed to address…”.
-  (if you’re writing a more general essay, you still need to summarize the source material you’re drawing on, so you can pull the same trick – explain in a few words what the general message in the source material is, but expressed in terms of what’s missing)
+  (if you’re writing a more general essay, you still need to summarize the source material you’re drawing on, so you can pull the same trick – explain in one sentence what the general message in the source material is, but expressed in terms of what’s missing)
 
   \emph{Explain, in one sentence, how you tackled the research question.}
   What’s your big new idea?
@@ -636,11 +645,11 @@ Kurzfassung der Arbeit.
   Did you run experiments?
   Build a piece of software?
   Carry out case studies?
-  This is likely to be the longest sentence, especially if it’s a PhD thesis – after all you’re probably covering several years worth of research.
+  This tends to be the longest sentence, above all in a PhD thesis, which covers years of research.
   But don’t overdo it – we’re still looking for a sentence that you could read aloud without having to stop for breath.
   Remember, the word ‘abstract’ means a summary of the main ideas with most of the detail left out.
   So feel free to omit detail!
-  (For those of you who got this far and are still insisting on writing an essay rather than signing up for a PhD, this sentence is really an elaboration of sentence 4 – explore the consequences of your new perspective).
+  (For those of you who got this far and are still insisting on writing an essay rather than signing up for a PhD, this sentence elaborates on sentence 4 – explore the consequences of your new perspective).
 
   \emph{As a single sentence, what’s the key impact of your research?
     Here we’re not looking for the outcome of an experiment.
@@ -776,7 +785,7 @@ A scientific argument always has the following steps that all must come in this
   \item[SM6] \emph{Now interpret your results} by contextualizing the measurements and drawing conclusion that lead to answering your research problem or defining further follow-up research problems.
 \end{itemize}
 %
-This template will mark various parts of the structure with SM1-SM6 to recall to you which step of a scientific argument is used and where.
+This template marks the parts of the structure with SM1-SM6 to recall to you which step of a scientific argument is used and where.
 
 \emph{Purpose and scope of <%- cref %>{sec:introduction}}.
 The introduction chapter is a summary of your work and your scientific argument that must be understandable to anyone in your scientific field, e.g., anyone in Data Science.
@@ -789,14 +798,14 @@ Section~\ref{sec:introduction:results} then presents the results and their inter
 A reader should need to read further only if they are not convinced or need more details to reproduce your study.
 The individual chapters and sections provide the details for each of the steps in your scientific argument.
 
-You usually write the introduction chapter \emph{after} you wrote all other chapters, but you should keep on making notes for each of the sections as you write the later chapters.
+Write the introduction chapter \emph{after} you wrote all other chapters, but you should keep on making notes for each of the sections as you write the later chapters.
 .
 
 \emph{Purpose and scope of the introduction paragraph to a chapter}.
 The paragraph you are reading above is a typical introductory paragraph to a chapter.
 It is a high-level summary of the chapters' topic (SM1 and SM2).
 It gives the reader some guidance by breaking down the chapter topic into subtopics that are named (SM3) in the right order with forward references to the corresponding sections (SM4).
-It may close with announcing the result you obtain (SM6) but this is usually not done in the opening paragraph of the introduction.
+It may close with announcing the result you obtain (SM6) but the opening paragraph of the introduction is not the place for it.
 
 \emph{Zwei Arten von Lesenden}.
 Die Arbeit soll sich von vorne bis hinten lesen lassen: Jeder Begriff wird eingeführt, bevor er verwendet wird.
@@ -832,7 +841,7 @@ This introduction has to
 \end{itemize}
 
 You have to provide citations/literature references for each of the statements and claims you are making.
-This section is usually a summary of the related work discussion in <%- cref %>{sec:background}.
+This section summarizes the related work discussion in <%- cref %>{sec:background}.
 
 At the end of this section, you have established a \emph{knowledge gap} between the state of the art and the general objective you developed in <%- cref %>{sec:introduction:topic}.
 \emph{Stating a (knowledge) gap between a status quo and a desired situation is the \emph{first step} of a writing scientific argument.}
@@ -843,7 +852,7 @@ At the end of this section, you have established a \emph{knowledge gap} between
 
 \emph{Purpose and scope}.
 In this section you state in which way you will address the knowledge gap you identified at the end of <%- cref %>{sec:introduction:state-of-art}.
-You usually cannot address and resolve the entire knowledge gap in your work.
+One work cannot address and resolve the entire knowledge gap.
 The purpose of this section is to detail the specific part of the knowledge that you will address.
 You thereby make all the assumptions explicit that underlie your work (``in this report we focus on genomes of female giraffes who lived in the years 1950-2000 in South Africa'').
 
@@ -854,7 +863,7 @@ Your general research question states
   \item and the evaluation criteria that will determine whether you are successful (``...that are present in at least 75\% of the studied giraffes'')
 \end{itemize}
 
-You will usually break your general research question down into sub-research questions.
+Break your general research question down into sub-research questions.
 You may do this here.
 The sub-research questions have to form a chain that take you in smaller steps from the starting point/assumptions of your general research question to your final objective and evaluation.
 
@@ -876,7 +885,7 @@ In each paragraph you first briefly explain the concept of the step in your meth
 
 \emph{Purpose and scope}.
 You close the introduction by stating the evaluation setup you designed to evaluate the success of your study regarding the research objective, which comes in two steps.
-It is most likely a summary of your evaluation in <%- cref %>{sec:evaluation}.
+It summarizes your evaluation in <%- cref %>{sec:evaluation}.
 
 <%= heading2 %>*{Results (SM5)}
 
@@ -885,9 +894,9 @@ You state the evaluation method that is in line with your research question from
 <%= heading2 %>*{Interpretation (SM6)}
 
 You summarize your interpretation of the results and draw conclusions.
-State whether and to which degree the research question from <%- cref %>{sec:introduction:research-question} has been answered successfully or not.
+State whether, and to which degree, the research question from <%- cref %>{sec:introduction:research-question} has been answered.
 
-Finally state briefly how much closer society and science have come in answering the general objective you outlined in <%- cref %>{sec:introduction:topic}.
+Close by stating how much closer society and science have come in answering the general objective you outlined in <%- cref %>{sec:introduction:topic}.
 
 %%% ===============================================================================
 \section{Background (SM1)}\label{sec:background}
@@ -952,7 +961,7 @@ provide details based on <%- cref %>{sec:problem-exposition:context-understandin
 
 \emph{Purpose and scope}.
 After you stated research context (SM1), research problem (SM2), and research method (SM3) in <%- cref %>{sec:introduction} and possibly <%- cref %>{sec:problem-exposition}, the remainder of your entire report addresses execution (SM4), results (SM5), and interpretation (SM6).
-You usually do this by addressing various sub-problems again through scientific arguments following the 6 steps SM1-SM6.
+Do this by addressing the sub-problems through scientific arguments that follow the 6 steps SM1-SM6.
 
 Have a short chapter introduction that recalls and explains the first research problem of your thesis.
 The problem has to show up in the introduction in <%- cref %>{sec:introduction:research-question} or in <%- cref %>{sec:problem-exposition:research-problems} already.
@@ -980,7 +989,7 @@ if certain details take too much space.
 
 \emph{The third, fourth, and following paragraph provides details on applying the method or developing a new approach, i.e., execution (SM4) and may explain results (SM5)}, i.e. details on the steps needed to reproduce the results.
 
-Results (SM5) can come in many forms, e.g., conceptual diagrams, algorithms, tables, charts, a list of articles from a literature research etc. You must reference them (``<%- cref %>{fig:my_label} shows.
+Results (SM5) come in different forms, e.g., conceptual diagrams, algorithms, tables, charts, a list of articles from a literature research etc. You must reference them (``<%- cref %>{fig:my_label} shows.
 ..'') and describe the results in text.
 If you use diagrams, tables, or charts, you cannot expect the reader to know what to you expect them to see in a diagram, table or chart.
 Describe to them how to read these, explain the meaning of particular elements, point out special observations.
@@ -998,7 +1007,7 @@ Make use of appendices if certain details take too much space.
 Here you can infer what a particular observation means (for you), how it can be applied, or what others can do with it.
 Do not interpret your results before you have described all of them.
 This is a common mistake done by most beginner writers.
-You want to quickly get to the point, which is the final finding or interpretation.
+Get to the point: the final finding or interpretation.
 But you forget that your reader does not understand yet what you are interpreting - they do not know yet what you do know.
 Interpret the results only after describing all of them.
 The interpretation must be based on the written description only.
@@ -1129,6 +1138,10 @@ Weitere Hinweise gibt \href{https://doi.org/10.1007/978-1-4471-6639-9}{\emph{Wri
 {\LTGendlineexample=\numexpr\inputlineno-1\relax%
   \tcbinputlisting{%
     listing only,
+<% if (twocolumn) { -%>
+    % Über Spalten und Seiten umbrechen. Eine nicht umbrechbare Box ragt in die Gleitobjekte oder den unteren Rand, wenn die Gleitobjekte einer Seite den schmalen Spalten zu wenig Platz lassen
+    breakable,
+<% } -%>
     listing file=\currfilepath,
     colback=green!5!white,
     colframe=green!25,
@@ -1251,7 +1264,7 @@ See \url{https://help.jabref.org/en/CleanupEntries} for an overview of the clean
 <%= heading1 %>{Zusammenfassung und Ausblick}\label{sec:conclusion}
 %%% ===============================================================================
 
-Your conclusions are not just a factual summary of your work, but they position, interpret, and defend your findings against the state of the art that you discussed in <%- cref %>{sec:introduction:state-of-art}.
+Your conclusions are more than a factual summary of your work: they position, interpret, and defend your findings against the state of the art that you discussed in <%- cref %>{sec:introduction:state-of-art}.
 You specifically outline which concrete findings or methodological contributions advance our knowledge towards the general objective you introduced in <%- cref %>{sec:introduction:topic}.
 Objectively discuss which parts you solved and in which parts you failed.
 
@@ -1343,7 +1356,7 @@ Hier bitte einen kurzen Durchgang durch die Arbeit.
 <% break; default: -%>
 <% break; } -%>
 <% if (isPaper) { -%>
-<%= indent %>Identification of funding sources and other support, and thanks to individuals and groups that assisted in the research and the preparation of the work should be included in an acknowledgment section, which is placed just before the reference section in your document \cite{acmart}.
+<%= indent %>Identification of funding sources and other support, and thanks to individuals and groups that assisted in the research and the preparation of the work should be included in an acknowledgment section, which comes right before the reference section in your document \cite{acmart}.
 <% if (howtotext) { -%>
 
 <%= indent %>The hints on writing a paper/thesis are taken from Dirk Fahland's \href{https://github.com/dfahland/Master-or-Bachelor-thesis-Template-Eindhoven-University-of-Technology}{LaTeX template for Bachelor and Master theses at Eindhoven University of Technology}.
@@ -1442,6 +1455,41 @@ Alle Links wurden zuletzt am 29.03.2021 geprüft.
 
 \lipsum[1]
 
+%%% ===============================================================================
+<%= heading1 %>{Nutzung von KI-Tools}\label{sec:ai-tools}
+%%% ===============================================================================
+
+% Hier wird dokumentiert, inwiefern KI-Tools zur Erstellung wesentlicher Teile dieser Arbeit verwendet wurden.
+<% if (thesisvariant == "ustutt") { -%>
+% Die Erklärung am Ende dieser Arbeit fordert diesen Anhang.
+% Siehe die Formvorgaben für Abschlussarbeiten des Fachbereichs Informatik der Universität Stuttgart: https://www.f05.uni-stuttgart.de/informatik/dokumente/Formulare/Formvorgaben-Abschlussarbeiten.pdf
+<% } -%>
+% Die Struktur ist mit der Prüferin oder dem Prüfer abzustimmen.
+% Nur übernommene KI-Outputs (z.B. Text, Abbildungen) müssen aufgeführt werden; die Nutzung von KI zur Recherche oder Rechtschreibprüfung nicht.
+% Wurden keine KI-Outputs als wesentliche Teile übernommen, ist dies statt der Tabelle zu erklären.
+
+<% if (examples) { -%>
+\begin{tabular}{lll}
+  \toprule
+  KI-Tool          & Genutzt für                    & Genutzt in                \\
+  \midrule
+  ChatGPT5         & Erstellung von Abbildungen     & <%- cref %>{fig:label,fig:jpg} \\
+  Google AI search & Pseudocode für Algorithmus XYZ & <%- cref %>{alg:example}       \\
+  DeepL Write      & Umformulierung der Einleitung  & <%- cref %>{sec:introduction}  \\
+  \bottomrule
+\end{tabular}
+<% } else { -%>
+\begin{tabular}{lll}
+  \toprule
+  KI-Tool          & Genutzt für                    & Genutzt in               \\
+  \midrule
+  ChatGPT5         & Erstellung von Abbildungen     & Abb. 3.4, 5.2 und 6.3    \\
+  Google AI search & Pseudocode für Algorithmus XYZ & Listing 4.1              \\
+  DeepL Write      & Umformulierung der Einleitung  & <%- cref %>{sec:introduction} \\
+  \bottomrule
+\end{tabular}
+<% } -%>
+
 \pagestyle{empty}
 \renewcommand*{\chapterpagestyle}{empty}
 <% if (documentclass == "scientific-thesis") { -%>

package/generators/app/templates/main.en.tex

@@ -55,9 +55,18 @@
 % DO NOT DOWNLOAD IEEEtran.cls - Use the one of your LaTeX distribution
 \documentclass[<%= ieeevariant %>,<%= papersize %>paper,english]{IEEEtran}[2015/08/26]
 
-% Balance the last page using the balance package (see https://ctan.org/pkg/balance)
-% Alternative to balance is the pbalance package (see https://ctan.org/pkg/pbalance), which sometimes works better
-\usepackage{balance}
+% Balance the columns on the last page using the flushend package (see https://ctan.org/pkg/flushend).
+% It needs no command in the text.
+% The balance package (see https://ctan.org/pkg/balance) needs \balance in the first column of the last page, otherwise it warns "You have called \balance in second column".
+% The pbalance package (see https://ctan.org/pkg/pbalance) does not work together with stfloats, which is loaded for \fnbelowfloat.
+% IEEEtran's own way is \IEEEtriggeratref, see the bibliography section at the end of this file.
+\usepackage{flushend}
+% flushend activates its balancing output routine only right before the final \clearpage, which is too late in some documents (e.g., the German example paper under LuaLaTeX): the last page is shipped out before. Thus, \flushcolsend activates it already here.
+% flushend only balances when the text of the last page reaches the second column.
+% If the text ends in the first column, end that column and put an empty box into the second one, so that flushend distributes the text over both columns.
+\makeatletter
+\AtEndDocument{\flushcolsend\if@firstcolumn\ifdim\pagetotal>\z@\newpage\hbox to\columnwidth{}\fi\fi}
+\makeatother
 <% break; case "scientific-thesis": -%>
 
 % The following package allows \\ at the title page
@@ -279,7 +288,7 @@
 <% if (useExampleEnvironment) { -%>
 \usepackage[realmainfile]{currfile}
 \usepackage{tcolorbox}
-\tcbuselibrary{<%= listings %>}
+\tcbuselibrary{<%= listings %><% if (twocolumn) { %>,breakable<% } %>}
 <% } -%>
 <% } -%>
 
@@ -578,7 +587,7 @@ Zweitgutachter: Dr. Eduard Exempel               %TODO Bei Bedarf Geschlechter d
 In one sentence, what’s the topic?
 Phrase it in a way that your reader will understand.
 If you’re writing a PhD thesis, your readers are the examiners – assume they are familiar with the general field of research, so you need to tell them specifically what topic your thesis addresses.
-Same advice works for scientific papers – the readers are the peer reviewers, and eventually others in your field interested in your research, so again they know the background work, but want to know specifically what topic your paper covers.
+Same advice works for scientific papers – the readers are the peer reviewers, and later others in your field interested in your research, so again they know the background work, but want to know specifically what topic your paper covers.
 
 \emph{State the problem you tackle.}
 What’s the key research question?
@@ -589,12 +598,12 @@ If you can’t summarize your thesis/paper/essay in one key question, then you d
 Keep working at this step until you have a single, concise (and understandable) question.
 
 \emph{Summarize (in one sentence) why nobody else has adequately answered the research question yet.}
-For a PhD thesis, you’ll have an entire chapter, covering what’s been done previously in the literature.
+For a PhD thesis, you’ll have an entire chapter, covering the prior work in the literature.
 Here you have to boil that down to one sentence.
-But remember, the trick is not to try and cover all the various ways in which people have tried and failed; the trick is to explain that there’s this one particular approach that nobody else tried yet (hint: it’s the thing that your research does).
+But remember, the trick is not to try and cover all the ways in which people have tried and failed; the trick is to explain that there’s this one particular approach that nobody else tried yet (hint: it’s the thing that your research does).
 But here you’re phrasing it in such a way that it’s clear it’s a gap in the literature.
 So use a phrase such as “previous work has failed to address…”.
-(if you’re writing a more general essay, you still need to summarize the source material you’re drawing on, so you can pull the same trick – explain in a few words what the general message in the source material is, but expressed in terms of what’s missing)
+(if you’re writing a more general essay, you still need to summarize the source material you’re drawing on, so you can pull the same trick – explain in one sentence what the general message in the source material is, but expressed in terms of what’s missing)
 
 \emph{Explain, in one sentence, how you tackled the research question.}
 What’s your big new idea?
@@ -605,11 +614,11 @@ What’s your overall view on the question you introduced in step 2?)
 Did you run experiments?
 Build a piece of software?
 Carry out case studies?
-This is likely to be the longest sentence, especially if it’s a PhD thesis – after all you’re probably covering several years worth of research.
+This tends to be the longest sentence, above all in a PhD thesis, which covers years of research.
 But don’t overdo it – we’re still looking for a sentence that you could read aloud without having to stop for breath.
 Remember, the word ‘abstract’ means a summary of the main ideas with most of the detail left out.
 So feel free to omit detail!
-(For those of you who got this far and are still insisting on writing an essay rather than signing up for a PhD, this sentence is really an elaboration of sentence 4 – explore the consequences of your new perspective).
+(For those of you who got this far and are still insisting on writing an essay rather than signing up for a PhD, this sentence elaborates on sentence 4 – explore the consequences of your new perspective).
 
 \emph{As a single sentence, what’s the key impact of your research?
   Here we’re not looking for the outcome of an experiment.
@@ -750,7 +759,7 @@ A scientific argument always has the following steps that all must come in this
   \item[SM6] \emph{Now interpret your results} by contextualizing the measurements and drawing conclusion that lead to answering your research problem or defining further follow-up research problems.
 \end{itemize}
 %
-This template will mark various parts of the structure with SM1-SM6 to recall to you which step of a scientific argument is used and where.
+This template marks the parts of the structure with SM1-SM6 to recall to you which step of a scientific argument is used and where.
 
 \emph{Purpose and scope of <%- cref %>{sec:introduction}}.
 The introduction chapter is a summary of your work and your scientific argument that must be understandable to anyone in your scientific field, e.g., anyone in Data Science.
@@ -763,14 +772,14 @@ Section~\ref{sec:introduction:results} then presents the results and their inter
 A reader should need to read further only if they are not convinced or need more details to reproduce your study.
 The individual chapters and sections provide the details for each of the steps in your scientific argument.
 
-You usually write the introduction chapter \emph{after} you wrote all other chapters, but you should keep on making notes for each of the sections as you write the later chapters.
+Write the introduction chapter \emph{after} you wrote all other chapters, but you should keep on making notes for each of the sections as you write the later chapters.
 .
 
 \emph{Purpose and scope of the introduction paragraph to a chapter}.
 The paragraph you are reading above is a typical introductory paragraph to a chapter.
 It is a high-level summary of the chapters' topic (SM1 and SM2).
 It gives the reader some guidance by breaking down the chapter topic into subtopics that are named (SM3) in the right order with forward references to the corresponding sections (SM4).
-It may close with announcing the result you obtain (SM6) but this is usually not done in the opening paragraph of the introduction.
+It may close with announcing the result you obtain (SM6) but the opening paragraph of the introduction is not the place for it.
 
 \emph{Two kinds of readers}.
 Write your report so that it can be read from beginning to end: introduce every concept before you use it.
@@ -806,7 +815,7 @@ This introduction has to
 \end{itemize}
 
 You have to provide citations/literature references for each of the statements and claims you are making.
-This section is usually a summary of the related work discussion in <%- cref %>{sec:background}.
+This section summarizes the related work discussion in <%- cref %>{sec:background}.
 
 At the end of this section, you have established a \emph{knowledge gap} between the state of the art and the general objective you developed in <%- cref %>{sec:introduction:topic}.
 \emph{Stating a (knowledge) gap between a status quo and a desired situation is the \emph{first step} of a writing scientific argument.}
@@ -817,7 +826,7 @@ At the end of this section, you have established a \emph{knowledge gap} between
 
 \emph{Purpose and scope}.
 In this section you state in which way you will address the knowledge gap you identified at the end of <%- cref %>{sec:introduction:state-of-art}.
-You usually cannot address and resolve the entire knowledge gap in your work.
+One work cannot address and resolve the entire knowledge gap.
 The purpose of this section is to detail the specific part of the knowledge that you will address.
 You thereby make all the assumptions explicit that underlie your work (``in this report we focus on genomes of female giraffes who lived in the years 1950-2000 in South Africa'').
 
@@ -828,7 +837,7 @@ Your general research question states
   \item and the evaluation criteria that will determine whether you are successful (``...that are present in at least 75\% of the studied giraffes'')
 \end{itemize}
 
-You will usually break your general research question down into sub-research questions.
+Break your general research question down into sub-research questions.
 You may do this here.
 The sub-research questions have to form a chain that take you in smaller steps from the starting point/assumptions of your general research question to your final objective and evaluation.
 
@@ -850,7 +859,7 @@ In each paragraph you first briefly explain the concept of the step in your meth
 
 \emph{Purpose and scope}.
 You close the introduction by stating the evaluation setup you designed to evaluate the success of your study regarding the research objective, which comes in two steps.
-It is most likely a summary of your evaluation in <%- cref %>{sec:evaluation}.
+It summarizes your evaluation in <%- cref %>{sec:evaluation}.
 
 <%= heading2 %>*{Results (SM5)}
 
@@ -859,9 +868,9 @@ You state the evaluation method that is in line with your research question from
 <%= heading2 %>*{Interpretation (SM6)}
 
 You summarize your interpretation of the results and draw conclusions.
-State whether and to which degree the research question from <%- cref %>{sec:introduction:research-question} has been answered successfully or not.
+State whether, and to which degree, the research question from <%- cref %>{sec:introduction:research-question} has been answered.
 
-Finally state briefly how much closer society and science have come in answering the general objective you outlined in <%- cref %>{sec:introduction:topic}.
+Close by stating how much closer society and science have come in answering the general objective you outlined in <%- cref %>{sec:introduction:topic}.
 
 %%% ===============================================================================
 \section{Background (SM1)}\label{sec:background}
@@ -926,7 +935,7 @@ provide details based on <%- cref %>{sec:problem-exposition:context-understandin
 
 \emph{Purpose and scope}.
 After you stated research context (SM1), research problem (SM2), and research method (SM3) in <%- cref %>{sec:introduction} and possibly <%- cref %>{sec:problem-exposition}, the remainder of your entire report addresses execution (SM4), results (SM5), and interpretation (SM6).
-You usually do this by addressing various sub-problems again through scientific arguments following the 6 steps SM1-SM6.
+Do this by addressing the sub-problems through scientific arguments that follow the 6 steps SM1-SM6.
 
 Have a short chapter introduction that recalls and explains the first research problem of your thesis.
 The problem has to show up in the introduction in <%- cref %>{sec:introduction:research-question} or in <%- cref %>{sec:problem-exposition:research-problems} already.
@@ -954,7 +963,7 @@ if certain details take too much space.
 
 \emph{The third, fourth, and following paragraph provides details on applying the method or developing a new approach, i.e., execution (SM4) and may explain results (SM5)}, i.e. details on the steps needed to reproduce the results.
 
-Results (SM5) can come in many forms, e.g., conceptual diagrams, algorithms, tables, charts, a list of articles from a literature research etc. You must reference them (``<%- cref %>{fig:my_label} shows.
+Results (SM5) come in different forms, e.g., conceptual diagrams, algorithms, tables, charts, a list of articles from a literature research etc. You must reference them (``<%- cref %>{fig:my_label} shows.
 ..'') and describe the results in text.
 If you use diagrams, tables, or charts, you cannot expect the reader to know what to you expect them to see in a diagram, table or chart.
 Describe to them how to read these, explain the meaning of particular elements, point out special observations.
@@ -972,7 +981,7 @@ Make use of appendices if certain details take too much space.
 Here you can infer what a particular observation means (for you), how it can be applied, or what others can do with it.
 Do not interpret your results before you have described all of them.
 This is a common mistake done by most beginner writers.
-You want to quickly get to the point, which is the final finding or interpretation.
+Get to the point: the final finding or interpretation.
 But you forget that your reader does not understand yet what you are interpreting - they do not know yet what you do know.
 Interpret the results only after describing all of them.
 The interpretation must be based on the written description only.
@@ -1071,7 +1080,7 @@ The remainder of the paper starts with a presentation of related work (<%- cref
 <% } -%>
 <% if (!examples && (documentclass == 'ieee')) { %>Afterwards, we present some Lorem Ipsum (<%- cref %>{sec:loremipsum}).
 <% } -%>
-Finally, a conclusion is drawn and outlook on future work is made (<%- cref %>{sec:outlook}).
+The paper closes with a conclusion and an outlook on future work (<%- cref %>{sec:outlook}).
 
 <%= heading1 %>{Related Work}
 \label{sec:relatedwork}
@@ -1097,6 +1106,10 @@ More hints are given in \href{https://doi.org/10.1007/978-1-4471-6639-9}{\emph{W
 {\LTGendlineexample=\numexpr\inputlineno-1\relax%
   \tcbinputlisting{%
     listing only,
+<% if (twocolumn) { -%>
+    % Split across columns and pages. An unbreakable box overflows into the floats or the bottom margin when the floats of a page leave the narrow columns too little room
+    breakable,
+<% } -%>
     listing file=\currfilepath,
     colback=green!5!white,
     colframe=green!25,
@@ -1221,7 +1234,7 @@ See \url{https://help.jabref.org/en/CleanupEntries} for an overview of the clean
 <%= heading1 %>{Conclusion and Outlook}\label{sec:conclusion}
 %%% ===============================================================================
 
-Your conclusions are not just a factual summary of your work, but they position, interpret, and defend your findings against the state of the art that you discussed in <%- cref %>{sec:introduction:state-of-art}.
+Your conclusions are more than a factual summary of your work: they position, interpret, and defend your findings against the state of the art that you discussed in <%- cref %>{sec:introduction:state-of-art}.
 You specifically outline which concrete findings or methodological contributions advance our knowledge towards the general objective you introduced in <%- cref %>{sec:introduction:topic}.
 Objectively discuss which parts you solved and in which parts you failed.
 
@@ -1311,7 +1324,7 @@ Be specific in the sense that your arguments for future work should be based on
 <% break; default: -%>
 <% break; } -%>
 <% if (isPaper) { -%>
-<%= indent%>Identification of funding sources and other support, and thanks to individuals and groups that assisted in the research and the preparation of the work should be included in an acknowledgment section, which is placed just before the reference section in your document \cite{acmart}.
+<%= indent%>Identification of funding sources and other support, and thanks to individuals and groups that assisted in the research and the preparation of the work should be included in an acknowledgment section, which comes right before the reference section in your document \cite{acmart}.
 <% if (howtotext) { -%>
 
 <%= indent%>The hints on writing a paper/thesis are taken from Dirk Fahland's \href{https://github.com/dfahland/Master-or-Bachelor-thesis-Template-Eindhoven-University-of-Technology}{LaTeX template for Bachelor and Master theses at Eindhoven University of Technology}.
@@ -1409,6 +1422,41 @@ All links were last followed on October 5, 2020.
 
 \lipsum[1]
 
+%%% ===============================================================================
+<%= heading1 %>{Use of AI Tools}\label{sec:ai-tools}
+%%% ===============================================================================
+
+% Document to what extent AI tools were used for generating significant parts of this work.
+<% if (thesisvariant == "ustutt") { -%>
+% The declaration at the end of this thesis requires this appendix.
+% See the layout requirements of the CS department of the University of Stuttgart: https://www.f05.uni-stuttgart.de/informatik/dokumente/Formulare/Formvorgaben-Abschlussarbeiten_en.pdf
+<% } -%>
+% Agree on its structure with your examiner.
+% Only AI outputs (e.g., text or figures) adopted into this thesis need to be listed; using AI for searching or spell checking does not.
+% If no AI outputs were adopted as significant parts of this thesis, state this instead of the table.
+
+<% if (examples) { -%>
+\begin{tabular}{lll}
+  \toprule
+  AI Tool          & Used for                      & Used where                \\
+  \midrule
+  ChatGPT5         & Generating figures            & <%- cref %>{fig:label,fig:jpg} \\
+  Google AI search & Pseudo code for algorithm XYZ & <%- cref %>{alg:example}       \\
+  DeepL Write      & Rephrasing the introduction   & <%- cref %>{sec:introduction}  \\
+  \bottomrule
+\end{tabular}
+<% } else { -%>
+\begin{tabular}{lll}
+  \toprule
+  AI Tool          & Used for                      & Used where               \\
+  \midrule
+  ChatGPT5         & Generating figures            & Fig. 3.4, 5.2, and 6.3   \\
+  Google AI search & Pseudo code for algorithm XYZ & Listing 4.1              \\
+  DeepL Write      & Rephrasing the introduction   & <%- cref %>{sec:introduction} \\
+  \bottomrule
+\end{tabular}
+<% } -%>
+
 \pagestyle{empty}
 \renewcommand*{\chapterpagestyle}{empty}
 <% if (documentclass == "scientific-thesis") { -%>

package/generators/app/templates/minted.example.en.tex

@@ -48,3 +48,7 @@ public class Hello {
   \label{lst:flJava}
 \end{listing}
 <%- eexample %>
+<% if (isThesis) { -%>
+
+<%- include('minted.long.example.en.tex', this); -%>
+<% } -%>

package/generators/app/templates/minted.long.example.en.tex

@@ -0,0 +1,75 @@
+Listings longer than the remaining space continue on the next page if they are \emph{not} floating.
+A floating listing is always placed on a single page.
+Thus, use \verb+\captionof{listing}+ instead of the \verb+listing+ environment.
+<%- Cref %>{lst:long} spans multiple pages.
+
+\begin{minted}[linenos=true]{java}
+public class Squares {
+
+    /** Returns the square of 1. */
+    public int square1() {
+        return 1 * 1;
+    }
+
+    /** Returns the square of 2. */
+    public int square2() {
+        return 2 * 2;
+    }
+
+    /** Returns the square of 3. */
+    public int square3() {
+        return 3 * 3;
+    }
+
+    /** Returns the square of 4. */
+    public int square4() {
+        return 4 * 4;
+    }
+
+    /** Returns the square of 5. */
+    public int square5() {
+        return 5 * 5;
+    }
+
+    /** Returns the square of 6. */
+    public int square6() {
+        return 6 * 6;
+    }
+
+    /** Returns the square of 7. */
+    public int square7() {
+        return 7 * 7;
+    }
+
+    /** Returns the square of 8. */
+    public int square8() {
+        return 8 * 8;
+    }
+
+    /** Returns the square of 9. */
+    public int square9() {
+        return 9 * 9;
+    }
+
+    /** Returns the square of 10. */
+    public int square10() {
+        return 10 * 10;
+    }
+
+    /** Returns the square of 11. */
+    public int square11() {
+        return 11 * 11;
+    }
+
+    /** Returns the square of 12. */
+    public int square12() {
+        return 12 * 12;
+    }
+
+    public static void main(String[] args) {
+        System.out.println(new Squares().square12());
+    }
+}
+\end{minted}
+\captionof{listing}{Long Java listing spanning multiple pages}
+\label{lst:long}

package/generators/app/templates/otherfeatures.example.en.tex

@@ -16,7 +16,7 @@ Brackets work as designed:
 One can also input backticks in verbatim text: \verb|`test`|.
 <%- eexample %>
 
-Acronyms can be typeset in running text with \verb+\initialism+, which sets them as small caps that blend into the surrounding lowercase text. It is the lightweight alternative to the full acronym and glossary management (\verb+\gls+, with an automatic list of abbreviations) provided by the thesis variants: \verb+\initialism+ only typesets the acronym and does not add it to any list. The macro and a few ready-made acronyms (\verb+\OMG+, \verb+\BPEL+, \verb+\BPMN+, \verb+\UML+) are defined in \texttt{commands.tex}, where you can add your own.
+Acronyms can be typeset in running text with \verb+\initialism+, which sets them as small caps that blend into the surrounding lowercase text. It is the lightweight alternative to the full acronym and glossary management (\verb+\gls+, with an automatic list of abbreviations) provided by the thesis variants: \verb+\initialism+ only typesets the acronym and does not add it to any list. The macro and the ready-made acronyms (\verb+\OMG+, \verb+\BPEL+, \verb+\BPMN+, \verb+\UML+) are defined in \texttt{commands.tex}, where you can add your own.
 
 <%- bexample %>
 The \UML is maintained by the \OMG; related standards include \BPEL and \BPMN.

package/generators/app/templates/paragraphs.example.de.tex

@@ -17,7 +17,7 @@ Dies verwendet man quasi nie.
 
 Folglich werden neue Absätze insbesondere \emph{nicht} durch Doppelbackslashes erzeugt.
 Beispielsweise begann der letzte Satz in einem neuen Absatz.
-Eine ausführliche Motivation hierfür findet sich in \url{http://loopspace.mathforge.org/HowDidIDoThat/TeX/VCS/#section.3}.
+Andrew Stacey begründet dies ausführlich im Abschnitt <%- bquote %>Line Wrapping<%- equote %> seines \href{https://loopspace.mathforge.org/HowDidIDoThat/TeX/VCS/#section.3}{\emph{Document Revision System}}\footnote{\url{https://loopspace.mathforge.org/HowDidIDoThat/TeX/VCS/}}.
 <%- eexample %>
 
 <% if (documentclass == 'scientific-thesis') { -%>

package/generators/app/templates/paragraphs.example.en.tex

@@ -14,9 +14,9 @@ This leads to a hard line break.
 The text starts at the beginning of a new line.
 In LaTeX, you can do that by using two backslashes (\textbackslash\textbackslash).
 \\
-This is rarely used.
+Running text seldom needs a hard line break.
 
 Please do \textit{not} use two backslashes for new paragraphs.
 For instance, this sentence belongs to the same paragraph, whereas the last one started a new one.
-A long motivation for that is provided at \url{http://loopspace.mathforge.org/HowDidIDoThat/TeX/VCS/#section.3}.
+Andrew Stacey gives a long motivation for that in the section <%- bquote %>Line Wrapping<%- equote %> of his \href{https://loopspace.mathforge.org/HowDidIDoThat/TeX/VCS/#section.3}{\emph{Document Revision System}}\footnote{\url{https://loopspace.mathforge.org/HowDidIDoThat/TeX/VCS/}}.
 <%- eexample %>

package/generators/app/templates/subfig.example.en.tex

@@ -1,10 +1,11 @@
 <%# \subfloat is defined by the subfig package -%>
+<%# IEEE: figure* at the top ([!t]) as in IEEEtran's bare_conf.tex. LaTeX cannot place a figure* at the bottom; stfloats (loaded for \fnbelowfloat) can, but with "!" it may leave the columns too little room. -%>
 <%= heading2 %>{Sub Figures}
 
 An example of two sub figures is shown in <%- cref %>{fig:two_sub_figures}.
 
 <%- bexample %>
-\begin{figure<% if (documentclass == 'ieee') { %>*<% } %>}[!b]
+\begin{figure<% if (documentclass == 'ieee') { %>*<% } %>}[<%= documentclass == 'ieee' ? '!t' : '!b' %>]
   \centering
   \subfloat[Case I]{\includegraphics[width=.4\linewidth]{example-image-a}%
     \label{fig:first_case}}
@@ -27,7 +28,7 @@ e.g., \verb+\subfloat[]+.
 An example is shown in <%- cref %>{fig:two_sub_figures_ieee}.
 
 <%- bexample %>
-\begin{figure*}[!b]
+\begin{figure*}[!t]
   \centering
   \subfloat[]{\includegraphics[width=.4\linewidth]{example-image-a}%
     \label{fig:first_case_ieee}}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "generator-latex-template",
-  "version": "2026.9.28",
+  "version": "2026.10.8",
   "description": "Generates a latex template to be used as start for writing scientific works (papers, articles, theses)",
   "homepage": "https://github.com/latextemplates/generator-latex-template",
   "author": {