Saved in:
Bibliographic Details
Main Author: Nuriyev, Mahammad
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.01720
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915593819521024
author Nuriyev, Mahammad
author_facet Nuriyev, Mahammad
contents 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/
format Preprint
id arxiv_https___arxiv_org_abs_2511_01720
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Efficient Tool-Calling Multi-Expert NPC Agent for Commonsense Persona-Grounded Dialogue
Nuriyev, Mahammad
Computation and Language
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/
title Efficient Tool-Calling Multi-Expert NPC Agent for Commonsense Persona-Grounded Dialogue
topic Computation and Language
url https://arxiv.org/abs/2511.01720