You Have Been LaTeXpOsEd: A Systematic Analysis of Information Leakage in Preprint Archives Using Large Language Models

TL;DR

This work tackles the risk of information leakage from LaTeX source files in preprint archives by introducing LaTeXpOsEd, a four‑stage framework that combines traditional pattern matching and large language models (LLMs) to detect hidden disclosures across $100{,}000$ arXiv submissions (~$1.2$ TB). It also presents LLMSec-DB, a benchmark for evaluating 25 LLMs on secret‑detection tasks, and demonstrates that LLMs can uncover real credentials, PII, and internal communications that conventional methods miss, albeit at higher cost. The study delivers a large‑scale audit, revealing thousands of PII leaks, sensitive credentials, and confidential communications, and provides policy recommendations for authors and platforms to mitigate these risks while supporting open science. The findings underscore the need for automated sanitization, integrated credential scanning, and author guidance at submission to reduce preventable disclosures in public research artifacts and protect researchers and institutions.

Abstract

The widespread use of preprint repositories such as arXiv has accelerated the communication of scientific results but also introduced overlooked security risks. Beyond PDFs, these platforms provide unrestricted access to original source materials, including LaTeX sources, auxiliary code, figures, and embedded comments. In the absence of sanitization, submissions may disclose sensitive information that adversaries can harvest using open-source intelligence. In this work, we present the first large-scale security audit of preprint archives, analyzing more than 1.2 TB of source data from 100,000 arXiv submissions. We introduce LaTeXpOsEd, a four-stage framework that integrates pattern matching, logical filtering, traditional harvesting techniques, and large language models (LLMs) to uncover hidden disclosures within non-referenced files and LaTeX comments. To evaluate LLMs' secret-detection capabilities, we introduce LLMSec-DB, a benchmark on which we tested 25 state-of-the-art models. Our analysis uncovered thousands of PII leaks, GPS-tagged EXIF files, publicly available Google Drive and Dropbox folders, editable private SharePoint links, exposed GitHub and Google credentials, and cloud API keys. We also uncovered confidential author communications, internal disagreements, and conference submission credentials, exposing information that poses serious reputational risks to both researchers and institutions. We urge the research community and repository operators to take immediate action to close these hidden security gaps. To support open science, we release all scripts and methods from this study but withhold sensitive findings that could be misused, in line with ethical principles. The source code and related material are available at the project website https://github.com/LaTeXpOsEd

Figures (3)

• Figure 1: The LaTeXpOsEd framework: a four-step process for scraping, parsing, mining, and analyzing documents.
• Figure 2: Ratio of papers without usable comments in LaTeX source files.
• Figure 3: The ten most commonly occurring domains in URLs extracted from the comments.