发表机构
University of California, Berkeley; California Polytechnic State University; Hofstra University（加州大学伯克利分校; 加州州立理工大学; 霍夫斯特拉大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对专家级金融问答的合规与防幻觉需求，提出自改进RAG框架，通过三个专门智能体及反馈自校正实现，在FinanceBench数据集上获86%神谕引导准确率，为金融监管应用提供可解释方案。
AI 中文摘要
专家级金融问答既需要基于事实的验证以捕捉数值幻觉，又需要审计追踪以满足监管合规要求，这是标准的单次检索增强生成（RAG）系统所欠缺的特性。我们通过自改进RAG（Self-Improving RAG）框架向该目标迈进，该框架将文档问答分解为三个专门智能体（检索、推理和裁决），由编排器通过反馈驱动的自校正进行协调。当裁决智能体对答案的评分低于动态阈值时，系统会触发重试并采用升级策略：更广泛的检索、更谨慎的提示以及更宽松的接受标准。我们在FinanceBench（美国证券交易委员会文件问答）数据集上进行评估，自改进RAG实现了86%的神谕引导准确率（衡量与标准答案的一致性），并达到36.4%的拉撒路率（Lazarus Rate），通过针对性重试恢复了近十分之四的初始错误答案。一项关键发现是，带有裁决驱动重试的固定检索管道无需动态路由即可实现优异结果，并具备完全可解释性；每个决策都记录了置信度分数，为受监管的金融应用提供了所需的审计追踪。
英文摘要
Expert-level financial question answering requires both grounded verification to catch numeric hallucinations and audit trails for regulatory compliance, attributes that standard single-pass RAG systems lack. We take a step toward this goal with Self-Improving RAG, a framework that decomposes document QA into three specialized agents (Retrieval, Reasoning, and Judge) coordinated by an orchestrator with feedback-driven self-correction. When the Judge Agent scores an answer below a dynamic threshold, the system triggers retry with escalated strategies: broader retrieval, more careful prompting, and relaxed acceptance criteria. We evaluate on FinanceBench (SEC filing QA), where Self-Improving RAG achieves 86% oracle-guided accuracy (measuring agreement with gold answers) with a 36.4% Lazarus Rate, recovering nearly 4 in 10 initially incorrect answers through targeted retry. A key finding is that a fixed retrieval pipeline with judge-driven retry achieves strong results without dynamic routing, providing full interpretability. Every decision is logged with confidence scores, enabling the audit trails required for regulated financial applications.
Comments17 pages, 2 figures. Accepted at the ICLR 2026 Workshop on Advances in Financial AI
