动态拆解查询+自我反思，让大模型查得更准更省时
Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval
- 用智能代理拆分复杂问题，逐轮检索增强答案
- 运维领域整体得分提升0.04，平均倒数排名提高0.17
- 适合复杂结构化查询，多跳推理任务需权衡延迟
检索增强生成（RAG）通过引入外部知识提升大语言模型表现，但传统单步静态检索难以应对复杂查询。本文提出一种代理协同的自适应RAG框架，包含动态查询分解、迭代检索和有限自我反思评估循环。在两个互补数据集上评估：特定领域的运维知识库与多跳推理基准MuSiQue。使用总体得分、引用准确率、平均倒数排名和主题覆盖度等指标发现，查询分解在结构化领域带来稳定提升（总体得分+0.04，MRR+0.17），但在多跳任务中降低排序精度；而反思机制虽提升引用准确率，却显著增加延迟。结果表明，智能增强并非普遍适用，应根据查询与领域特征选择性应用。研究主张采用自适应、成本敏感的协同调度，而非统一激进的推理流程。
原文摘要 · Abstract (English)
Retrieval-Augmented Generation (RAG) enhances Large Language Models (LLMs) by grounding their responses in external knowledge, but conventional pipelines rely on static, single-step retrieval that limits performance on complex queries. This paper presents an Agent-Orchestrated Adaptive RAG framework that introduces dynamic query decomposition, iterative retrieval, and a bounded self-reflective evaluation loop. We evaluate the system across two complementary datasets: a domain-specific DevOps knowledge base and the multi-hop reasoning benchmark MuSiQue. Using metrics that include overall score, citation accuracy, mean reciprocal rank, and topic coverage, we find that query decomposition yields consistent gains in the structured domain (overall score $+0.04$, MRR $+0.17$ on DevOps) but degrades ranking precision on the multi-hop benchmark, while the reflection mechanism improves citation accuracy at a substantial latency cost. These contrasting results show that agentic enhancements are not universally beneficial and must be applied selectively according to query and domain characteristics. Our findings argue for adaptive, cost-aware orchestration rather than uniformly aggressive reasoning pipelines.
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。