Controlled Yet Natural: A Hybrid BDI-LLM Conversational Agent for Child Helpline Training

2025-09-20

AI Summary

Training child helpline counselors has long been hindered by the high cost of human role-playing and the linguistic rigidity of rule-based agents. This paper introduces a novel virtual dialogue agent that deeply integrates large language models (LLMs) with the Belief-Desire-Intention (BDI) framework: LLMs are innovatively embedded into three core BDI processes—intention recognition, response generation, and bypass regulation—thereby preserving behavioral controllability while substantially enhancing linguistic diversity and conversational realism. Script-based evaluation and within-subject experiments demonstrate that the system’s response quality is non-inferior to manually authored content (p > 0.05); moreover, users perceive it as significantly more credible, hold more positive attitudes toward it, and exhibit stronger engagement intentions (posterior probability = 0.845). This work establishes a new paradigm for generative training agents that are high-fidelity, interpretable, and dynamically controllable.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Conversational AI/Dialog Systems
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• 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

Abstract

Child helpline training often relies on human-led roleplay, which is both time- and resource-consuming. To address this, rule-based interactive agent simulations have been proposed to provide a structured training experience for new counsellors. However, these agents might suffer from limited language understanding and response variety. To overcome these limitations, we present a hybrid interactive agent that integrates Large Language Models (LLMs) into a rule-based Belief-Desire-Intention (BDI) framework, simulating more realistic virtual child chat conversations. This hybrid solution incorporates LLMs into three components: intent recognition, response generation, and a bypass mechanism. We evaluated the system through two studies: a script-based assessment comparing LLM-generated responses to human-crafted responses, and a within-subject experiment (N=37) comparing the LLM-integrated agent with a rule-based version. The first study provided evidence that the three LLM components were non-inferior to human-crafted responses. In the second study, we found credible support for two hypotheses: participants perceived the LLM-integrated agent as more believable and reported more positive attitudes toward it than the rule-based agent. Additionally, although weaker, there was some support for increased engagement (posterior probability = 0.845, 95% HDI [-0.149, 0.465]). Our findings demonstrate the potential of integrating LLMs into rule-based systems, offering a promising direction for more flexible but controlled training systems.

Problem

• Developing realistic virtual child chat conversations for helpline training
• Overcoming limited language understanding in rule-based training agents
• Integrating LLMs into BDI framework for controlled yet natural interactions

Innovation

• Integrates LLMs into BDI framework for realism
• Uses LLMs for intent recognition and response
• Hybrid system balances control with natural conversation