Efficient Tool-Calling Multi-Expert NPC Agent for Commonsense Persona-Grounded Dialogue
This addresses the problem of building efficient, persona-grounded NPC agents for interactive environments, representing an incremental improvement using existing methods.
The paper tackled creating NPCs for natural dialogue and action execution in interactive environments, achieving second place in the Commonsense Persona-Grounded Dialogue Challenge 2025 with a computationally efficient multi-expert system.
We present a multi-expert system for creating Non-Player Characters (NPCs) capable of both natural dialogue and contextual action execution in interactive environments. Using Qwen3 as the base model and Low-Rank Adaptation (LoRA) adapters, we instantiate three specialists: tool calling, tool-response interpretation, and direct dialogue. Our system comfortably meets the computational efficiency requirements, delivering fast responses and maintaining modest resource usage on L40S GPUs. In the Commonsense Persona-Grounded Dialogue Challenge 2025, our method ranked second overall. Code available at: https://github.com/MahammadNuriyev62/CPDC-challenge-2025-solution/