社区开发者基于 MiniCPM 构建可溯源文档校验 agent TraceCell
国内开源圈一个很实用的项目：用 MiniCPM 三个模型分工，把 OCR 识别结果做成带坐标和置信度的证据台账，可疑字段自动转人工。做票据、合同、财务单据自动化的开发者可以参考这套可追溯设计。
社区开发者 @mucerx 基于 OpenBMB 的 MiniCPM 系列开源了 TraceCell，一个本地优先的文档解析与校验 agent。系统由三个模型协作：MiniCPM-OCR 1.0 负责解析文档内容，MiniCPM-OCR Jev 给识别候选打分，MiniCPM-V 4.7 复核模糊区域，其答案再次打分后才被接受。
每个字段都通过证据台账关联到源图坐标、置信度和复核状态，所有结果可回溯到原图位置。不确定的结果会触发定向重解析、模型复核或人工介入，文档按自动接受、抽检、全量复核三类路由。TraceCell 还能做算术检查和单据对账，即使文字识别正确也能发现数字不一致。
作者当前的 demo 覆盖 5 份合成文档、83 个区域、52 个业务字段，在 78 个标注样本上校准出 80.77% 的自动接受率且无误接受，但这是小样本集结果，未经独立生产验证。系统支持 MCP 和 Skill，是把视觉理解变成可追溯决策的社区实践示例。
📄 Can your document agent show why a result should be trusted and flag what needs another look?
Community developer @mucerx built TraceCell, a local-first document parsing and verification agent powered by MiniCPM. 🔍
Three models work together: 📖 MiniCPM-OCR 1.0 parses document content. ⚖️ MiniCPM-OCR Jev scores recognition candidates. 🧠 MiniCPM-V 4.7 reviews ambiguous regions, with its answers scored again before acceptance.
📍 Every field links back to source-image coordinates, confidence, and review status in an evidence ledger.
🔄 Uncertain results trigger targeted re-parsing, model review, or human escalation. Documents are routed for automatic acceptance, spot checks, or full review.
🧮 TraceCell also checks arithmetic and reconciles documents—catching mismatches even when the text itself was recognized correctly.
🧪 The author’s current demo covers 5 synthetic documents, 83 regions, and 52 business fields. Calibration on 78 labelled samples reports 80.77% automatic acceptance with no incorrect acceptances. These are sample-set results, not independently verified production guarantees.
With MCP and Skill support, it’s a practical community example of turning visual understanding into traceable decisions. 💙
🛠️ Explore TraceCell: https://t.co/FECX9swHog
🚀 Build with MiniCPM: GitHub — MiniCPM-OCR: https://t.co/qG1294sFYs GitHub — MiniCPM-V: https://t.co/iYDxpa52tn 🤗 Hugging Face — MiniCPM-OCR 1.0: https://t.co/dBVKzHuCC6
#MiniCPM #OpenBMB #DocumentAI #OpenSource
This report is categorized under AITOP's "Products" desk, focusing on China's AI industry and domestic foundation model ecosystem. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include MiniCPM, OpenBMB, TraceCell, MCP/工具.
Source attribution: curated from OpenBMB (@OpenBMB).
