Deriving Character Logic from Storyline as Codified Decision Trees

ACL ARR 2026 January Submission1205 Authors

Keywords: Role-playing, Persona Modeling, Grounding System

Abstract:

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 \textbf{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, indicating that codified and validated behavioral representations lead to more reliable agent grounding.

Paper Type: Long

Research Area: Dialogue and Interactive Systems

Research Area Keywords: grounded dialog

Languages Studied: English

Submission Number: 1205