<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Evaluation on Human in the Loop</title><link>http://aitip.tech/en/tags/evaluation/</link><description>Recent content in Evaluation on Human in the Loop</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Mon, 16 Nov 2026 08:00:00 +0800</lastBuildDate><atom:link href="http://aitip.tech/en/tags/evaluation/index.xml" rel="self" type="application/rss+xml"/><item><title>94% of Fields Correct. Now What? A Two-Stage Architecture for Scanned Documents</title><link>http://aitip.tech/en/posts/two-stage-document-extraction/</link><pubDate>Mon, 16 Nov 2026 08:00:00 +0800</pubDate><guid>http://aitip.tech/en/posts/two-stage-document-extraction/</guid><description>Across 38 scans and 4,732 field assertions, the classic OCR pipeline passed 94.08% of fields, but only six documents passed every check. Page transcription, extraction and large-schema experiments show how to separate reading from interpretation, retain source evidence, and avoid treating model agreement as proof of correctness.</description></item></channel></rss>