Higher-Order Extraction from Unsiloed：从文档数值生成代码算出未印刷的答案
Unsiloed 这招挺实用：让 AI 从文档读数后写代码算出没印在纸上的数字，还能和原文核对置信度。
Higher-Order Extraction from Unsiloed 不只是定位页面上的数值，而是根据文档中的原始数值生成确定性代码并运行，计算出文档从未印刷的最终数字。每个结果都能追溯到页面来源。如果文档恰好印有该答案，系统会将计算结果与之比对，匹配则置信度上升，不匹配则置信度降为零。
Most document AI is built to answer one question: where is this number on the page?
But the number people need often isn't on the page. It has to be calculated from the numbers that are.
Higher-Order Extraction from Unsiloed does that calculation. It calculates numbers a document never prints by generating deterministic code from page values.
e.g. you ask for the final figure, it reads the raw values from the document, writes code that combines them, and runs that code to get the answer. Every result traces back to the page.
There's a second check. If the document happens to print the answer somewhere, the system compares it with its own result. If they match, confidence goes up. If they don't, confidence drops to zero.
