Interactive AI NPCs Powered by LLMs: Technical Report for the CPDC Challenge 2025

TL;DR

This technical report details a unified framework for improving AI-powered NPCs using large language models, combining context engineering and reinforcement learning to enhance dialogue stability, role-playing, and task performance in the CPDC 2025 challenge.

Contribution

The paper introduces a novel framework integrating context engineering and reinforcement learning, achieving top rankings in multiple challenge tracks for AI NPCs.

Findings

Achieved top rankings in CPDC 2025 challenge tracks.

Improved tool call stability and role-playing guidance.

Enhanced task-oriented dialogue performance through RL.

Abstract

This report presents the solution and results of our team MSRA\_SC in the Commonsense Persona-Grounded Dialogue Challenge (CPDC 2025). We propose a simple yet effective framework that unifies improvements across both GPU Track and API Track. Our method centers on two key components. First, Context Engineering applies dynamic tool pruning and persona clipping for input compression, combined with post-processing techniques such as parameter normalization and function merging. Together with manually refined prompts, this design improves tool call stability, execution reliability, and role-playing guidance. Second, in the GPU Track, we further adopt GRPO training, replacing supervised fine-tuning with reinforcement learning directly optimized by reward signals. This mitigates small-sample overfitting and significantly enhances task-oriented dialogue performance. In the final evaluation, our team ranks 1st in Task 2 API, 2nd in Task 1 API, and 3rd in both Task 3 API and GPU track, demonstrating the effectiveness of our approach. Our code is publicly available at https://gitlab.aicrowd.com/nikoo_yu/cpdc-2025-winning-solution