HyperWorld: Hypergraph-Structured State Serialization Improves Learned Textual World Models

HyperWorld：超图结构状态序列化提升学习型文本世界模型

Abstract: World models enable language-model agents to predict environment dynamics and plan before acting. In text environments, the model must learn symbolic action effects from serialized state descriptions, but the role of serialization structure remains underexplored. 摘要： 世界模型使语言模型智能体能够在行动前预测环境动态并进行规划。在文本环境中，模型必须从序列化的状态描述中学习符号化的动作效果，但序列化结构的作用目前尚未得到充分探索。

We present HyperWorld, a controlled study of state serialization for learned textual world models. We compare raw observations with three symbolic serializations of the same ground-truth state: independent sentences, pairwise triples, and entity-centered hyperedge units that group multiple related facts around entities and relations. 我们提出了 HyperWorld，这是一项针对学习型文本世界模型状态序列化的对照研究。我们将原始观测结果与同一真实状态的三种符号化序列表示进行了比较：独立句子、成对三元组，以及以实体为中心的超边单元（将多个相关事实围绕实体和关系进行分组）。

All variants use the same training objective: given a state and an action, predict symbolic effects or judge the action infeasible. Across model scales, data budgets, and in-distribution and out-of-distribution test worlds, hyperedge serialization gives the clearest gains for 0.5B—1.5B models and under distribution shift. 所有变体均使用相同的训练目标：给定一个状态和一个动作，预测符号化效果或判断该动作是否可行。在不同的模型规模、数据预算以及分布内和分布外的测试环境中，超边序列化在 0.5B 到 1.5B 参数规模的模型以及分布偏移的情况下，表现出了最显著的增益。

Larger models reduce the gap, and pairwise triples can match or slightly exceed hyperedges on in-distribution exact match, but hyperedges achieve the strongest out-of-distribution fact F1 and the best small-to-medium scale trade-off between feasibility detection and effect prediction. 更大的模型缩小了这种差距，且成对三元组在分布内精确匹配上可以达到或略微超过超边，但超边在分布外事实 F1 指标上表现最强，并在可行性检测与效果预测之间实现了中小型规模下的最佳权衡。

In downstream greedy planning, the hyperedge world model also attains the highest success rate among the tested representations. These results show that higher-order state organization is a simple but effective inductive bias for learned symbolic world models, especially when model capacity is limited or test environments differ from training. 在下游的贪婪规划任务中，超边世界模型在所有测试的表示方法中也获得了最高的成功率。这些结果表明，高阶状态组织对于学习型符号世界模型而言是一种简单但有效的归纳偏置，尤其是在模型容量有限或测试环境与训练环境存在差异时。