RefScout

Guides

Practical reading on citations, bibliographies, and the LaTeX submission workflow — written for people who actually submit papers.

Why your AI assistant keeps making up citations

Hallucinated references look exactly like real ones. Why language models invent citations, how often it happens, and where the line between "AI helping" and "AI inventing" actually is.

3 min read

The .bib errors that quietly wreck a submission

Duplicate entries under two keys, cite commands pointing at deleted entries, references nobody cites — the three .bib problems that slip past manual checks.

3 min read

A pre-submission checklist for IEEE and ACM papers

Page limits, double-blind anonymization slips, bibliography formatting — the small, boring requirements that get papers flagged on a technicality.

3 min read

Finding a real source for something you already know is true

You're sure the claim is true; you just can't remember where you read it. Why asking a chatbot backfires, and how to work from the claim to real candidate papers.

2 min read

\newacronym vs \gls — the short version

The LaTeX glossaries package in two commands: one defines, one uses. The compile errors that trip up new users, and how to retrofit acronyms into an existing paper.

2 min read

Comparing two drafts of a LaTeX paper without losing your mind

paper_v3.tex vs paper_v3_reviewer_comments.tex: when latexdiff is more friction than you need, a line-by-line diff answers "what actually changed" in two minutes.

2 min read

What counts as a "real" citation database, and where the gaps are

Semantic Scholar, OpenAlex, CrossRef, arXiv — what each covers, where each has gaps, and why "not found in one database" doesn't mean a paper is fake.

3 min read