Papers

CIEG框架提出证据治理:医疗智能体的工具输出需先审核再作答

Not All Evidence Is Admissible: Evidence Governance for Tool-Augmented Medical Agents

Why we picked it

做医疗AI或智能体的开发者该看看这篇。它指出了一个反直觉的问题:给模型加工具输出反而可能把答对的题做错。CIEG的卡片式准入方案不训练就能用,三个医疗任务上比同规模方案都准,成本还更低。

arXiv论文2611.22169指出工具增强医疗智能体的一个隐患:医学工具会静默失败,而冻结的推理模型无法忽略已进入上下文的内容,导致原本答对的案例在加入工具输出后被翻转答错,作者称之为证据劫持(evidence hijacking)。论文用 V-information 理论解释该现象:推理者失去忽略选项后,工具输出可能携带负可用信息,把筛选决策移到推理者之外才能恢复该选项。

Body

据此作者提出上下文无关证据治理框架CIEG:每个工具输出写成可与原始输入核对的证据卡片,一个独立检查器只看输入与卡片来决定哪些可准入,推理者仅基于准入卡片作答,全程无需训练。在医学问答、眼底青光眼筛查和胸片肺炎分类三个任务上,CIEG取得同等规模模型中的最高准确率,并在更低成本下显著超过基于同一骨干的最先进工具使用智能体和多智能体系统。

Original · arXiv cs.AI

Not All Evidence Is Admissible: Evidence Governance for Tool-Augmented Medical Agents

Tool-augmented medical agents treat whatever their tools return as evidence. We show that this default is costly: medical tools fail silently, and a frozen reasoner cannot ignore what enters its context, so many cases it would have answered correctly are flipped to wrong once tool outputs are added. We call this evidence hijacking. $\mathcal{V}$-information explains the phenomenon: the reasoner has lost the option to ignore, so a tool output can carry negative usable information, and moving the decision of what to keep outside the reasoner restores that option. The theory further shows that part of any filtering gain comes merely from letting the reasoner see less, and only selection by content lets it see better. Building on this, we recast tool-augmented reasoning as evidence governance, which separates deciding what counts as evidence from the reasoner that answers, and propose context-independent evidence governance (CIEG), the first medical agent framework that rules, after acquisition, on whether each individual tool output may enter as evidence. Every tool output is written as a card that can be checked against the raw input; an independent checker that sees only the input and the cards decides which cards become evidence; and the reasoner answers from the admitted cards alone, with no training anywhere. On medical question answering, fundus glaucoma screening, and chest X-ray pneumonia classification, it achieves the highest accuracy among models of comparable scale, and significantly outperforms state-of-the-art tool-using agents and multi-agent systems built on the same backbone at lower cost.

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Papers" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. 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 agents, medical AI, CIEG, tool use.

Source attribution: curated from arXiv cs.AI (@Runxun Zhang, Onofrio Antonio Catalano, Jinqiao Wang, Jingwei Wei).