Yikai Zhao, Saurabh Pandey, Pradeep Kumar Misra · 2026 · arXiv
DOI: 10.48550/arXiv.2610.30026 · open-access PDF
企业代理人在文件上采取行动. 在代理人能够遵循政策、回答合规性问题或提交票之前,某些吞吐层必须将 Word 文件或 PDF 转化为代币 - 这个层是在一个在野外部署的代理人悄悄地失去信息的地方。我们显示每个主流转换器失去一个层 - 这意味着作者在文本的外观中编码 - 而且损失使代理人自信地错了,而不是明显不完整。
Machine translation (Meta M2M100); the English original follows.
Enterprise agents act on documents. Before an agent can follow a policy, answer a compliance question or file a ticket, some ingestion layer must turn a Word file or PDF into tokens - and that layer is where an agent deployed in the wild silently loses information. We show that one layer is lost by every mainstream converter - the meaning an author encodes in how text looks - and that the loss makes agents confidently wrong rather than visibly incomplete. We build a task from 76 real policy documents: recover the requirements the author singled out, scored against the file's own OOXML properti
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.30026, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange