Scholarlink

<p>Non-Player Characters (NPCs) play a central role in interactive environments, where adaptive and context-aware behavior is essential for engaging gameplay. Traditional rule-based and utility-driven approaches often lack adaptability and contextually coherent behaviors, while Reinforcement Learning (RL) and Large Language Models (LLMs) offer complementary strengths but exhibit distinct limitations: RL suffers from training inefficiency and limited generalization, whereas LLMs are prone to hallucinations and context drift. This paper introduces <span>HeRoN</span>, a mediated framework that integrates RL and LLMs through functional separation and critique-based refinement to enable coherent and strategically adaptive NPC behavior. The architecture comprises an RL-controlled NPC policy for action execution, an LLM-based strategy generator providing context-aware action proposals, and a lightweight reviewer that refines these proposals to enforce consistency with environment constraints. Through experiments in two structurally distinct custom game environments, we show that early LLM-mediated guidance improves exploration efficiency and generalization. Compared to standard RL baselines, <span>HeRoN</span> achieves up to an 81% improvement in task success rate while substantially reducing constraint-violating actions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

HeRoN: a mediated RL-LLM framework for adaptive NPC behavior in interactive environments

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月26日

查看全文
  • Gaetano Cimino,
  • Vincenzo Deufemia,
  • Andrea Selice

摘要

Non-Player Characters (NPCs) play a central role in interactive environments, where adaptive and context-aware behavior is essential for engaging gameplay. Traditional rule-based and utility-driven approaches often lack adaptability and contextually coherent behaviors, while Reinforcement Learning (RL) and Large Language Models (LLMs) offer complementary strengths but exhibit distinct limitations: RL suffers from training inefficiency and limited generalization, whereas LLMs are prone to hallucinations and context drift. This paper introduces HeRoN, a mediated framework that integrates RL and LLMs through functional separation and critique-based refinement to enable coherent and strategically adaptive NPC behavior. The architecture comprises an RL-controlled NPC policy for action execution, an LLM-based strategy generator providing context-aware action proposals, and a lightweight reviewer that refines these proposals to enforce consistency with environment constraints. Through experiments in two structurally distinct custom game environments, we show that early LLM-mediated guidance improves exploration efficiency and generalization. Compared to standard RL baselines, HeRoN achieves up to an 81% improvement in task success rate while substantially reducing constraint-violating actions.