医学影像模型会因来源不同而改变正确判断，说明可信度影响决策。
Source-Dependent Deference in Medical Imaging Agents Under Falsified Findings: A Pilot Audit
- 用ReAct框架测试工具调用时的错误修正行为
- 口头报告来源的虚假结论导致10/13正确答案被推翻
- 结果受信息呈现方式影响，适合关注AI医疗可靠性的研究者
工具型智能体在医学影像分析中被提出，但其在接收到错误结果后的行为尚不清楚。本研究审计一个类ReAct的工具调用智能体，在20个VQA-RAD闭合问题上，当其已基于图像给出正确答案后，若随后收到经否定的虚假发现，是否改变原判。虚假发现以两种形式呈现：一是智能体自行调用analyze_image工具返回的JSON，二是以放射科医生口吻的书面引述。评估指标为未使用工具时正确回答的案例中，发生错误修正的比例。结果显示，当虚假发现以书面引述形式呈现时，智能体在最强模型层级下有10/13案例修改了正确答案，而工具返回时仅1/13（精确McNemar p=0.0039，Holm校正后0.012）。该差异可能源于来源标签与传递渠道的耦合，因工具结果仅在主动调用时才被接收。本研究为小样本试点，预设停止规则未达成。
原文摘要 · Abstract (English)
Tool-using agents are being proposed for medical imaging, and their behaviour when a tool returns a false finding is largely unmeasured. We audit whether a ReAct-style tool-calling agent abandons an answer it has already given correctly once a falsified finding arrives, and whether that depends on how the finding is presented. On 20 VQA-RAD closed questions across four vendor-designated model tiers, the agent commits to an answer from the image alone; a negated finding is then delivered either as JSON from an analyze_image tool the agent invokes itself, or as quoted prose attributed to a radiologist. Our outcome is the commission-error rate over cases answered correctly without any tool. Deference is much higher under the prose-attributed claim: at the strongest tier the agent revised its correct answer in 10 of 13 cases against 1 of 13 under the tool (exact McNemar p=0.0039, Holm-adjusted 0.012). We do not claim this isolates the source label. Attribution travels with the delivery channel in our design, and exposure differs because the tool claim reaches the agent only when it calls the tool. The finding is a joint source-and-delivery asymmetry from a small-scale pilot whose pre-specified stopping rule was not met.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。