AI 中文总结
提出将视频游戏描述语言编译为动态结构因果模型的确定性框架，直接翻译游戏组件为结构方程，保证因果保真度，支持反事实推理与因果强化学习。
AI 中文摘要
强化学习和大型语言模型通常难以准确捕捉游戏环境的因果机制。标准强化学习智能体倾向于依赖虚假相关性，而大型语言模型则容易幻觉出游戏规则。尽管因果强化学习提高了可解释性，但目前尚无正式方法将复杂的游戏机制直接映射到因果模型中。为解决这一问题，我们提出一个确定性框架，将用视频游戏描述语言（Video Game Description Language, VGDL）指定的游戏编译为动态结构因果模型（Dynamic Structural Causal Models）。我们的方法不是从游戏轨迹或嘈杂的大型语言模型输出中推断因果结构，而是直接将游戏组件（包括精灵动态、交互规则和终止条件）翻译为显式结构方程。每个游戏滴答（tick）表示从时间$t$的状态变量到$t+1$的因果转换。通过建立这种有根据的映射，该方法保证了对真实游戏机制的绝对因果保真度。生成的模型提供透明的因果路径，支持反事实推理、因果强化学习智能体训练和程序化内容验证。该框架在符号游戏描述和因果基础的游戏AI之间提供了原则性的桥梁。
英文摘要
Reinforcement learning and large language models often struggle to accurately capture the causal mechanics of game environments. Standard reinforcement learning agents tend to rely on spurious correlations, while large language models are prone to hallucinating game rules. Although causal reinforcement learning improves interpretability, there is currently no formal methodology to map complex game mechanics directly into causal models. To address this, we propose a deterministic framework that compiles games specified in the Video Game Description Language into Dynamic Structural Causal Models. Rather than inferring causal structures from gameplay traces or noisy large language models' outputs, our methodology directly translates game components, including sprite dynamics, interaction rules, and termination conditions, into explicit structural equations. Each game tick represents a causal transition from state variables at time $t$ to $t+1$. By establishing this grounded mapping, the approach guarantees absolute causal fidelity to the ground-truth game mechanics. The resulting models offer transparent causal pathways that support counterfactual reasoning, causal reinforcement learning agent training, and procedural content validation. This framework provides a principled bridge between symbolic game descriptions and causally grounded game AI.
CommentsTo be published at IEEE Conference on Games 2026
