提出检测并阻断多步推理中错误传播的框架，提升AI生成结果可靠性。
Cascading Hallucination in Agentic RAG: The CHARM Framework for Detection and Mitigation
- 在多步推理流程中引入四类错误传播模式，构建分阶段验证机制
- 在多个数据集上实现89.4%错误检测率，误报率仅5.3%，延迟增加不足200ms
- 适用于生产级AI系统，支持人工介入监控，保障可解释性与治理
多步代理检索增强生成（RAG）流水线在复杂推理任务中表现出强大能力，但对一类现有幻觉检测机制无法识别的失败模式仍存在脆弱性：级联幻觉，即早期阶段引入的错误会在后续推理步骤中传播并放大，导致看似自信但事实错误的最终输出。为此，我们首次将级联幻觉定义为代理RAG系统中独立的故障模式，提出四类级联模式分类，并引入CHARM（Cascading Hallucination Aware Resolution and Mitigation）框架，用于检测和中断错误传播。CHARM包含四个组件——阶段级事实验证、跨阶段一致性追踪、置信度传播监控与级联消解触发，可无缝集成至标准代理RAG流水线，无需替换架构。我们在HotpotQA、MuSiQue、2WikiMultiHopQA及自建对抗数据集上评估了基于LangChain的代理流水线配置，实现89.4%的级联检测率，5.3%的误报率，每阶段平均延迟开销为215毫秒±18毫秒，相比输出层检测器（仅18.5%）显著提升82.1%的错误传播抑制效果。组件消融实验表明各模块均对整体覆盖有实质性贡献。CHARM可与人机协同监督框架结合，构成生产级代理AI部署的完整可靠性与治理体系。
原文摘要 · Abstract (English)
Multi-step agentic retrieval-augmented generation (RAG) pipelines have demonstrated significant capability for complex reasoning tasks, yet remain vulnerable to a class of failure that existing hallucination detection mechanisms systematically miss: cascading hallucination, where errors introduced at early pipeline stages propagate and amplify across successive reasoning steps, producing confident but factually incorrect final outputs. To address this vulnerability, we formalize cascading hallucination as a distinct failure mode in agentic RAG systems, present a four-type taxonomy of cascade patterns, and introduce CHARM (Cascading Hallucination Aware Resolution and Mitigation), an architectural framework for detecting and interrupting error propagation in multi-step reasoning pipelines. CHARM comprises four components - stage-level fact verification, cross-stage consistency tracking, confidence propagation monitoring, and cascade resolution triggering - that operate alongside standard agentic RAG pipelines without requiring architectural replacement. We evaluate CHARM on HotpotQA, MuSiQue, 2WikiMultiHopQA, and a custom adversarial dataset across LangChain agentic pipeline configurations, achieving an 89.4% cascade detection rate with a 5.3% false positive rate and 215 ms +/- 18 ms average latency overhead per stage, achieving an error propagation reduction of 82.1%, compared to 18.5% for output-level detectors. Component ablations confirm that each detection module contributes meaningfully to overall cascade coverage. CHARM integrates with human-in-the-loop oversight frameworks to provide a complete reliability and governance stack for production agentic AI deployment.
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。