Reliable Collaborative Conversational Agent System Based on LLMs and Answer Set Programming

2025-05-09

AI Summary

Large language model (LLM)-driven task-oriented dialogue (TOD) suffers from unreliable knowledge grounding, unsafe multi-agent coordination, and vulnerability to adversarial information injection. Method: This paper proposes a dual-agent collaborative framework integrating LLMs with Answer Set Programming (ASP), introducing the novel “Administrator–Assistant” dual-ASP agent paradigm. Collaboration is governed by a Collaboration Rule Set (CRS) that enables declarative knowledge encapsulation, implicit inter-agent communication, and safety-governed coordination—eliminating risky explicit message passing while ensuring verifiable reasoning and controlled knowledge updates. Contribution/Results: The AutoManager system, instantiated from this framework, was deployed in a real-world Taco Bell drive-thru scenario. Empirical evaluation demonstrates significantly higher task completion accuracy and robustness compared to commercial AI ordering systems. This work presents the first empirical validation of ASP-augmented LLM collaboration for secure and reliable TOD in production environments.

Technology Category

• Multiagent Systems: Coordination and Collaboration
• Planning, Routing, and Scheduling: Planning with Language Models
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

As the Large-Language-Model-driven (LLM-driven) Artificial Intelligence (AI) bots became popular, people realized their strong potential in Task-Oriented Dialogue (TOD). However, bots relying wholly on LLMs are unreliable in their knowledge, and whether they can finally produce a correct result for the task is not guaranteed. The collaboration among these agents also remains a challenge, since the necessary information to convey is unclear, and the information transfer is by prompts -- unreliable, and malicious knowledge is easy to inject. With the help of logic programming tools such as Answer Set Programming (ASP), conversational agents can be built safely and reliably, and communication among the agents made more efficient and secure. We proposed an Administrator-Assistant Dual-Agent paradigm, where the two ASP-driven bots share the same knowledge base and complete their tasks independently, while the information can be passed by a Collaborative Rule Set (CRS). The knowledge and information conveyed are encapsulated and invisible to the users, ensuring the security of information transmission. We have constructed AutoManager, a dual-agent system for managing the drive-through window of a fast-food restaurant such as Taco Bell in the US. In AutoManager, the assistant bot takes the customer's order while the administrator bot manages the menu and food supply. We evaluated our AutoManager and compared it with the real-world Taco Bell Drive-Thru AI Order Taker, and the results show that our method is more reliable.

Problem

• Ensuring reliability in LLM-driven task-oriented dialogue systems
• Improving secure collaboration among conversational AI agents
• Preventing malicious knowledge injection in agent communication

Innovation

• Uses Answer Set Programming for reliable agents
• Implements Administrator-Assistant Dual-Agent paradigm
• Encapsulates knowledge via Collaborative Rule Set