Loud or Silent? A Reusable Framework for Per-Modality Failure Analysis in Multimodal Clinical AI

• 类型：arxiv
• 标识：2608.01462
• 链接：https://arxiv.org/abs/2608.01462
• 主分类：multimodal
• 形态：application
• 被引：0
• 被引来源：Semantic Scholar
• S2被引：0
• 影响力被引：0
• TLDR：A model-agnostic modality-failure framework that returns a per-example failure taxonomy, a per-modality complementarity matrix that attributes error to modalities, and a loud-vs-silent dropout profile separating monitorable failures from those that pass unflagged far from the decision boundary, using only deployment-observable signals.
• 待LLM分类：否
• 标题中文：标题 -> 标题中文：响亮还是沉默？面向多模态临床 AI 的可复用逐模态失效分析框架
• TLDR中文：一种模型无关的模态失败框架，仅依赖部署可观察信号，返回逐样本失败分类、逐模态互补矩阵（将错误归因到模态）以及响亮 vs 静默 dropout 画像（区分可监控失败与远离决策边界未被标记的失败）
• 来源文件：
• /inbox/tom/_candidates/2026-08-05-agent-rag-longcontext-candidates.json
• [S2 enrich]