发表机构
Mentomy AI（曼托米人工智能公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出一种可审计的符号-RAG-生成式架构，以GRACE引擎为核心，通过约束策略最大化业务进展并保障访客效用，在跨领域对话评估中实现高准确率与单调性。
AI 中文摘要
目标导向的对话系统必须能够回答事实性问题、理解访客提供的信息，并推进业务目标，同时避免沦为僵化的问卷。本文提出了一种以目标导向的检索增强对话引擎（GRACE）为核心的符号-RAG-生成式架构。一个受指令约束的业务目标编译器将业务意图转化为不可变的目标集、归一化优先级向量、规范问题以及初始状态向量。在运行时，GRACE接收完整的对话历史、最新的访客消息、当前状态以及由独立RAG组件生成的基于事实的答案。它仅根据访客提供的证据更新完成度，并选择一个受上下文调节的后续问题。核心策略在满足最小访客效用约束的前提下，最大化预期的业务进展。我们形式化了状态、单调转移、来源分离、问题调节和约束策略；给出了参考架构；并定义了一项评估，包含24段英语房地产和10段西班牙语专业清洁对话，共计119个协议定义的访客回合。在两个领域中，GRACE实现了84.9%的精确状态转移准确率、91.6%的证据精确率、89.6%的证据召回率、100%的单调性以及94.1%的终态准确率。该评估在标准、多目标、RAG绕行、验证、拒绝和鲁棒性场景中确立了令人信服的符号状态性能。
英文摘要
Goal-oriented conversational systems must answer factual questions, understand visitor-provided information, and advance business objectives without becoming rigid questionnaires. This paper proposes a Symbolic-RAG-Generative architecture centered on the Goal-oriented Retrieval-Augmented Conversation Engine (GRACE). An instruction-constrained Business Goal Compiler transforms business intent into an immutable objective set, normalized priority vector, canonical questions, and initial state vector. At runtime, GRACE receives the complete conversation history, latest visitor message, current state, and grounded answer generated by a separate RAG component. It updates completion only from visitor-authored evidence and selects one contextually modulated follow-up. The core policy maximizes expected business progress subject to a minimum visitor-utility constraint. We formalize the state, monotonic transitions, source separation, question modulation, and constrained policy; present the reference architecture; and define an evaluation comprising 24 English real-estate and 10 Spanish professional-cleaning conversations, totaling 119 protocol-defined visitor turns. Across both domains, GRACE achieves 84.9% exact state-transition accuracy, 91.6% evidence precision, 89.6% evidence recall, 100% monotonicity, and 94.1% terminal-state accuracy. The evaluation establishes compelling symbolic-state performance across standard, multi-goal, RAG-detour, validation, refusal, and robustness scenarios.
Comments14 pages, 3 figures, 6 tables, Appendices with complete experiments
