用类型化状态机管控复杂问答的检索流程，提升准确率与效率。
StateRAG: Typed State Contracts for Complex Retrieval-Augmented Generation
- 引入外部类型化状态机管理检索路径与决策，分工明确
- 在11项指标中10项领先，长文本任务最高提升9.8分F1
- 比ReAct更省模型调用，适合需要精准控制的复杂推理场景
复杂检索增强生成需控制证据检索、下一步检索路径、探索方向、证据充分性判断及中间结果保留。现有RAG方法通过特定模型上下文、遍历流程、验证信号和记忆实现这些决策。本文提出StateRAG，将检索控制作为外部类型化状态独立于最终阅读器。该状态记录查询计划、类型化遍历路径、候选证据、验证结论及可复用成果，具有明确定义的字段语义和指定更新来源。角色操作符按序提议字段值，控制器验证并提交接受的提案。一次紧凑证据检查可触发跳过（Bypass）。否则，控制器结合已用预算选择释放（Release）、修改（Revise）或回退（Fallback）。最终阅读器仅在终端动作后调用，且每查询最多一次。在评估的多种方法中，StateRAG在LongBench、QASPER和DocVQA上11项质量指标中有10项得分最高。其LongBench宏平均EM和F1分别优于ReAct 6.3和6.1个百分点，最大任务级提升达9.8 F1点（MuSiQue）。在标准化模型尺寸加权下，相比ReAct，StateRAG在所有五个工作负载上均实现更低的每查询平均大语言模型令牌消耗。在匹配的LongBench对照实验中，类型化、角色专属载体较自由形式共享载体在宏平均EM、F1和准确率上分别提升1.3、2.2和3.0个百分点，平均检索开销降低6.2%。移除TAM、MARS或SMP均导致所有报告的LongBench和QASPER消融指标下降。
原文摘要 · Abstract (English)
Complex retrieval-augmented generation requires evidence retrieval and control over what to retrieve next, which paths to explore, whether evidence is sufficient, and which intermediate results to retain. Existing RAG paradigms encode these decisions through method-specific model contexts, traversal procedures, verification signals, and memory. We introduce StateRAG, which represents retrieval control as a typed state external to the final reader. The state records the query plan, typed traversal path, candidate evidence, verification verdict, and reusable artifacts, with defined field semantics and designated update sources. Ordered role operators propose field values, and the controller validates and commits accepted proposals. A one-time compact-evidence check may select Bypass. Otherwise, the controller combines the committed verdict with the remaining budget to select Release, Revise, or Fallback. The final reader is invoked only after a terminal action and at most once per query. Among the evaluated methods, StateRAG achieves the best score on 10 of 11 reported quality metrics across LongBench, QASPER, and DocVQA. Its LongBench macro-average EM and F1 exceed ReAct by 6.3 and 6.1 percentage points, and its largest task-level gain is 9.8 F1 points on MuSiQue. Under normalized model-size weighting, StateRAG records lower mean LLM-token use per query than ReAct across all five workloads. In a matched LongBench control, the typed, role-owned carrier exceeds a free-form shared carrier by 1.3, 2.2, and 3.0 percentage points in macro EM, F1, and accuracy, with 6.2% lower mean RET. Removing TAM, MARS, or SMP lowers every reported LongBench and QASPER ablation metric.
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。