为何智能体缓存失效及如何修复：基于少样本学习的结构化意图规范化
Why Agent Caching Fails and How to Fix It: Structured Intent Canonicalization with Few-Shot Learning
AI总结：
本文提出结构化意图分解框架W5H2，通过少样本学习显著提升缓存效率，实验显示在多个基准测试中性能优于现有方法，同时实现成本降低。
摘要：
个人AI代理通过重复调用LLM造成显著成本。我们显示现有缓存方法失效：GPTCache在真实基准上达到37.9%的准确率；APC达到0-12%。根本原因是优化了错误的属性——缓存有效性需要键一致性和精度，而非分类准确率。我们观察到缓存键评估归结为聚类评估，并应用V-measure分解将这些分离于n=8,682点上MASSIVE、BANKING77、CLINC150和NyayaBench v2，我们的新8,514条多语言代理数据集（528意图，20 W5H2类别，63种语言）。我们引入W5H2，结构化意图分解框架。使用SetFit每类8个示例，W5H2在MASSIVE上达到91.1%±1.7%在约2ms内——对比GPTCache的37.9%和20B参数LLM在3,447ms的68.8%。在NyayaBench v2（20类）上，SetFit达到55.3%，跨30种语言的跨语言迁移。我们的五级级联处理85%的交互本地化，投影97.5%的成本降低。我们通过RCPS提供受风险控制的选择性预测保证，有九种边界族。
英文摘要：
Personal AI agents incur substantial cost via repeated LLM calls. We show existing caching methods fail: GPTCache achieves 37.9% accuracy on real benchmarks; APC achieves 0-12%. The root cause is optimizing for the wrong property -- cache effectiveness requires key consistency and precision, not classification accuracy. We observe cache-key evaluation reduces to clustering evaluation and apply V-measure decomposition to separate these on n=8,682 points across MASSIVE, BANKING77, CLINC150, and NyayaBench v2, our new 8,514-entry multilingual agentic dataset (528 intents, 20 W5H2 classes, 63 languages). We introduce W5H2, a structured intent decomposition framework. Using SetFit with 8 examples per class, W5H2 achieves 91.1%+/-1.7% on MASSIVE in ~2ms -- vs 37.9% for GPTCache and 68.8% for a 20B-parameter LLM at 3,447ms. On NyayaBench v2 (20 classes), SetFit achieves 55.3%, with cross-lingual transfer across 30 languages. Our five-tier cascade handles 85% of interactions locally, projecting 97.5% cost reduction. We provide risk-controlled selective prediction guarantees via RCPS with nine bound families.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
