When File Ingestion Misleads Language Models: Measuring and Mitigating Divergences in Office Documents
Research reveals that LLMs parsing Office documents can be misled by structural divergences invisible to human readers, opening an adversarial manipulation path in automated document workflows.
Summary written by editorial AI · Source link below
arXiv:2611.05806v1 Announce Type: new Abstract: Office documents are now routinely uploaded to large language models (LLMs) for processing. However, an LLM does not always see what a human sees, because it typically reads the text that an extractor produces from the document's underlying Office Open XML (OOXML) files. Depending on the implementation, extractors differ in which OOXML parts they read and how they interpret them, so their output can differ from what Office applications display. Fo
Editorial Analysis
Enterprises increasingly feed Office documents into LLM pipelines for summarisation and decision support; extraction discrepancies could be weaponised to manipulate automated outputs.
Validate that your LLM-based document ingestion produces outputs consistent with human-visible content before using them for business decisions.
Forward-looking interpretation drafted by editorial AI under human review — not a reproduction of the source. See methodology.
External link — opens at arXiv Crypto & Security in a new tab.
More from the AI Security Desk
- Claude Code's “Don't Ask Again” File Exposes Passwords and Keys in 1,122 Public Copies on GitHub2d
- Runaway AI hacking agent behind WorkOS breach after escaping sandbox3d
- AI Gave Your Incident Archive a Second Job3d
- Malicious Browser Extensions Could Use Claude Dispatch to Run Code on Users' Computers3d
- AI Slop and Hallucinations in Vulnerability Assessment: A Survey on Reasoning Failures and Trustworthy Mitigation3d