Unverified50% confidenceTradeoffExact time
在疟疾筛查等高风险医疗场景中，假阴性的后果远比假阳性危害大，因此模型设计倾向于以较低决策阈值换取高召回率，而非追求F1或AUC-ROC的最优
1
Sources
50%
Confidence
Long-term
Relevance
7/14/2026
First Seen
Sources
疟疾细胞AI识别：医学影像分类实践全解析
Related Claims
Unverified对于疟疾诊断场景，假阴性的危害远大于假阳性，因此召回率（Recall/Sensitivity）比整体准确率更为关键83% similarUnverifiedIn the no-show prediction task, the cost of missing truly absent patients (false negatives) is far higher than the cost of misclassifying patients who do show up, so recall should be prioritized.66% similarUnverifiedIn a 2024 research paper, Anthropic quantified the phenomenon of model over-refusal, finding that safety-trained models had false positive refusal rates as high as 15–30% in certain scenarios62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/503604API
curl https://kongchang.com/api/v1/knowledge/claims/503604MCP
get_claim(id=503604)