游戏对话的去扁平化：在基于大语言模型的NPC中平衡角色真实性与任务执行

摘要

大型语言模型（LLMs）的兴起为游戏环境中创建动态非玩家角色（NPCs）开辟了新机遇，不仅实现了功能性任务执行，还确保了角色一致性对话的生成。本文中，我们（Tu_Character_lab）报告了参与2025年第二回合常识性角色基础对话挑战赛（CPDC）的情况，该赛事评估了代理在三个赛道上的表现：任务导向对话、上下文感知对话及其整合。我们的方法结合了两种互补策略：(i) 在API赛道采用轻量级提示技术，包括一种去角色化提示方法，以抑制过度角色扮演并提升任务忠实度；(ii) 在GPU赛道微调大型模型，利用Qwen3-14B进行监督微调（SFT）及低秩适应（LoRA）。我们的最佳提交在任务1中排名第二，在任务3（API赛道）中排名第二，在任务3（GPU赛道）中排名第四。

English

The emergence of large language models (LLMs) has opened new opportunities for cre- ating dynamic non-player characters (NPCs) in gaming environments, enabling both func- tional task execution and persona-consistent dialogue generation. In this paper, we (Tu_Character_lab) report our participation in the Commonsense Persona-Grounded Dialogue Challenge (CPDC) 2025 Round 2, which eval- uates agents across three tracks: task-oriented dialogue, context-aware dialogue, and their integration. Our approach combines two complementary strategies: (i) lightweight prompting techniques in the API track, including a Deflanderization prompting method to suppress excessive role-play and improve task fidelity, and (ii) fine-tuned large models in the GPU track, leveraging Qwen3-14B with supervisedfinetuning (SFT) and Low-Rank Adaptation(LoRA). Our best submissions ranked 2nd on Task 1, 2nd on Task 3 (API track), and 4th on Task 3 (GPU track).

游戏对话的去扁平化：在基于大语言模型的NPC中平衡角色真实性与任务执行

Deflanderization for Game Dialogue: Balancing Character Authenticity with Task Execution in LLM-based NPCs

摘要

Support