具有模块化记忆的固定角色小语言模型：消费级硬件上的可扩展NPC对话系统
Fixed-Persona SLMs with Modular Memory: Scalable NPC Dialogue on Consumer Hardware
AI总结：
针对LLMs在游戏NPC对话中硬件需求高、延迟大等问题，提出基于微调SLMs和运行时可交换记忆模块的系统，在消费级硬件上实现可扩展对话，还可应用于虚拟助手等领域。
摘要：
大语言模型（LLMs）在生成类人文本方面展现出卓越能力，但在计算机游戏对话系统中的适用性仍受限。这一限制源于其高昂的硬件需求、延迟约束，以及在游戏场景中需维持清晰知识边界的必要性。本文提出一种模块化NPC对话系统，该系统利用小语言模型（SLMs），通过微调编码特定NPC角色，并集成运行时可交换的记忆模块。这些记忆模块保存角色特定的对话上下文和世界知识，在游戏过程中无需重新训练或重新加载模型即可实现富有表现力的交互和长期记忆。我们使用三个开源SLMs全面评估系统：DistilGPT-2、TinyLlama-1.1B-Chat和Mistral-7B-Instruct，它们在合成的角色对齐数据上训练，并在消费级硬件上进行基准测试。虽然我们的方法受游戏应用启发，但其模块化设计和角色驱动的记忆架构在需要富有表现力、可扩展且记忆丰富的对话智能体的更广泛领域（如虚拟助手、客户支持机器人或交互式教育系统）中具有巨大潜力。
英文摘要：
Large Language Models (LLMs) have demonstrated remarkable capabilities in generating human-like text, yet their applicability to dialogue systems in computer games remains limited. This limitation arises from their substantial hardware requirements, latency constraints, and the necessity to maintain clearly defined knowledge boundaries within a game setting. In this paper, we propose a modular NPC dialogue system that leverages Small Language Models (SLMs), fine-tuned to encode specific NPC personas and integrated with runtime-swappable memory modules. These memory modules preserve character-specific conversational context and world knowledge, enabling expressive interactions and long-term memory without retraining or model reloading during gameplay. We comprehensively evaluate our system using three open-source SLMs: DistilGPT-2, TinyLlama-1.1B-Chat, and Mistral-7B-Instruct, trained on synthetic persona-aligned data and benchmarked on consumer-grade hardware. While our approach is motivated by applications in gaming, its modular design and persona-driven memory architecture hold significant potential for broader adoption in domains requiring expressive, scalable, and memory-rich conversational agents, such as virtual assistants, customer support bots, or interactive educational systems.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
