{"slug": "beyond-the-editing-canvas-evidence-divergence-in-ooxml-to-llm-ingestion", "title": "Beyond the Editing Canvas: Evidence Divergence in Ooxml-to-LLM Ingestion", "summary": "A new arXiv preprint (2608.25880) reveals that Office Open XML (OOXML) documents can yield different evidentiary content when displayed in Microsoft Office versus when extracted for large language model (LLM) pipelines, a condition the authors call 'plural ground truth.' The researchers identified 21 'evidence forks' across Excel, Word, and PowerPoint, and found that in tests with four native-ingestion LLM APIs and seven web chatbots, the APIs returned task-relevant facts hidden from the Office view in 48–76% of trials, with at least one interface exposing the trap for 20 of 21 mechanisms.", "body_md": "# Computer Science > Software Engineering\n\n[Submitted on 26 Aug 2026]\n\n# Title:Beyond the Editing Canvas: Evidence Divergence in OOXML-to-LLM Ingestion\n\n[View PDF](/pdf/2608.25880)\n\n[HTML (experimental)](https://arxiv.org/html/2608.25880v1)\n\nAbstract:LLM pipelines increasingly ingest Office Open XML (OOXML) documents (Word, Excel, and PowerPoint files) as first-class evidence in financial, compliance, and retrieval-augmented workflows, implicitly assuming semantic integrity: that the evidence consumed by the model matches the content shown in the Microsoft Office suite editing canvas. We show that this assumption can fail in OOXML-to-LLM pipelines. The same specification-valid OOXML file can yield one evidentiary view in Microsoft Office and another when extracted for an LLM. Each view is treated as authoritative by its consumer, a condition we call plural ground truth. The ingestion contract rarely states which view and semantic roles become model evidence or preserves how that evidence was derived. We call the specification-grounded OOXML constructions that induce such divergence evidence forks.\n\nWe systematically traverse and mine the OOXML specification and confirm 21 evidence forks across Excel, Word, and PowerPoint, spanning six dimensions of view construction. All 13 tools in our extraction panel emit evidence from at least one fork. We test four native-ingestion LLM APIs and seven web chatbots. Each test document carries a trap: a task-relevant fact exposed by extraction but not shown in Office. Across this 21-mechanism evaluation, the four APIs return the trap in 48--76% of trials. For 20 of 21 mechanisms, at least one of the eleven interfaces returns the trap. Our measurements further show that exposure is shaped upstream of the model by the ingestion path and extractor configuration. A source-level survey of sixteen popular open-source LLM projects further shows that default OOXML ingestion paths concentrate on affected extractor families.\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/beyond-the-editing-canvas-evidence-divergence-in-ooxml-to-llm-ingestion", "canonical_source": "https://arxiv.org/abs/2608.25880", "published_at": "2026-08-29 06:07:07+00:00", "updated_at": "2026-08-29 06:18:10.387167+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-research", "ai-safety"], "entities": ["arXiv", "Microsoft Office", "Office Open XML"], "alternates": {"html": "https://wpnews.pro/news/beyond-the-editing-canvas-evidence-divergence-in-ooxml-to-llm-ingestion", "markdown": "https://wpnews.pro/news/beyond-the-editing-canvas-evidence-divergence-in-ooxml-to-llm-ingestion.md", "text": "https://wpnews.pro/news/beyond-the-editing-canvas-evidence-divergence-in-ooxml-to-llm-ingestion.txt", "jsonld": "https://wpnews.pro/news/beyond-the-editing-canvas-evidence-divergence-in-ooxml-to-llm-ingestion.jsonld"}}