从情节中推导人物逻辑：基于决策树的编码方法

摘要

角色扮演（RP）智能体依赖行为配置文件在不同叙事情境中保持行为一致性，但现有配置方案普遍存在非结构化、不可执行及验证薄弱的问题，导致智能体行为脆弱。我们提出编码化决策树（CDT）——一种从大规模叙事数据中推导可执行、可解释决策结构的数据驱动框架。CDT将行为配置表征为条件规则树：内部节点对应经过验证的场景条件，叶节点编码具体行为陈述，从而在执行时实现上下文适配规则的确定性检索。该框架通过迭代推导候选场景-动作规则、进行数据验证及层级细化学习决策树，最终生成支持透明检视与原则性更新的行为配置。在涵盖16个叙事作品的85个角色基准测试中，CDT显著优于人工编写配置及先前的配置推导方法，表明经过编码与验证的行为表征能够实现更可靠的智能体行为锚定。

English

Role-playing (RP) agents rely on behavioral profiles to act consistently across diverse narrative contexts, yet existing profiles are largely unstructured, non-executable, and weakly validated, leading to brittle agent behavior. We propose Codified Decision Trees (CDT), a data-driven framework that induces an executable and interpretable decision structure from large-scale narrative data. CDT represents behavioral profiles as a tree of conditional rules, where internal nodes correspond to validated scene conditions and leaves encode grounded behavioral statements, enabling deterministic retrieval of context-appropriate rules at execution time. The tree is learned by iteratively inducing candidate scene-action rules, validating them against data, and refining them through hierarchical specialization, yielding profiles that support transparent inspection and principled updates. Across multiple benchmarks, CDT substantially outperforms human-written profiles and prior profile induction methods on 85 characters across 16 artifacts, indicating that codified and validated behavioral representations lead to more reliable agent grounding.

从情节中推导人物逻辑：基于决策树的编码方法

Deriving Character Logic from Storyline as Codified Decision Trees

摘要

Support