This paper introduces the Definite Finite Automaton augmented large language
model (DFA-LLM), a novel framework designed to enhance the capabilities of
conversational agents using large language models (LLMs). Traditional LLMs face
challenges in generating regulated and compliant responses in special scenarios
with predetermined response guidelines, like emotional support and customer
service. Our framework addresses these challenges by embedding a Definite
Finite Automaton (DFA), learned from training dialogues, within the LLM. This
structured approach enables the LLM to adhere to a deterministic response
pathway, guided by the DFA. The advantages of DFA-LLM include an interpretable
structure through human-readable DFA, context-aware retrieval for responses in
conversations, and plug-and-play compatibility with existing LLMs. Extensive
benchmarks validate DFA-LLM's effectiveness, indicating its potential as a
valuable contribution to the conversational agent.

介绍了一种新颖的框架，Definite Finite Automaton augmented large language model (DFA-LLM)，旨在增强使用大规模语言模型（LLM）的对话代理的能力。该框架通过将来自训练对话的学习得来的 Definite Finite Automaton (DFA) 嵌入到 LLM 中，从而使 LLM 能够遵循确定性的响应路径，具备解释性结构、上下文感知的对话响应检索以及与现有 LLMs 的即插即用兼容性。广泛的基准测试验证了 DFA-LLM 的有效性，表明其在对话代理领域具有潜在的价值贡献。