发表机构
Bloomberg（彭博社）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对多智能体RAG系统各组件孤立优化问题，提出GRADRAG框架，通过将RAG流程建模为计算图，传播评估反馈更新上游智能体，在两种检索范式基准测试中优于单步细化基线，提升了系统性能。
AI 中文摘要
检索增强生成（RAG）系统越来越多地采用多个语言模型智能体。然而，大多数先前工作孤立地优化组件，而非协调整个流程的改进。我们引入GRADRAG，一个跨组件提示适应框架，将RAG流程建模为计算图，并传播结构化评估反馈以更新上游智能体。评估器批判下游答案和支持证据，生成可操作反馈，提示优化器据此迭代更新诸如检索器、图构造器和回答器等自适应智能体。评估器在输出令人满意时还会触发提前停止。我们在两种检索范式下的SQUALITY和QMSUM基准上评估GRADRAG：使用IRCoT风格查询细化的基于扁平块的检索，以及从文档构建并迭代丰富实体关系图的基于图的检索。在这两种设置下，GRADRAG始终优于仅更新最终生成器的单步细化基线，在语言模型判断的成对比较中实现了12 - 15个百分点的净偏好优势，且大部分增益在两次细化迭代内实现。
英文摘要
Retrieval-Augmented Generation (RAG) systems increasingly employ multiple LLM agents. Yet, most prior work optimizes components in isolation rather than coordinating improvements across the pipeline. We introduce GRADRAG, a framework for cross-component prompt adaptation that models the RAG pipeline as a computational graph and propagates structured evaluation feedback to update upstream agents. An Evaluator critiques downstream answers and supporting evidence, producing actionable feedback that a Prompt Optimizer uses to iteratively update adaptive agents, such as retrievers, graph constructors, and answerers. The Evaluator also triggers early stopping when the output is deemed satisfactory. We evaluate GRADRAG on the SQUALITY and QMSUM benchmarks under two retrieval paradigms: flat chunk-based retrieval using IRCoT-style query refinement (Trivedi et al., 2023), and graph-based retrieval that constructs and iteratively enriches an entity-relation graph from the document. Across both settings, GRADRAG consistently outperforms one-step refinement baselines that update only the final generator, achieving a 12-15 percentage point net preference margin in LLM-judged pairwise comparisons, with most gains realized within two refinement iterations.
Comments8 pages
