RELATE-Sim: Leveraging Turning Point Theory and LLM Agents to Predict and Understand Long-Term Relationship Dynamics through Interactive Narrative Simulations
Matthew Yue, Zhikun Xu, Vivek Gupta, Thao Ha, Liesal Sharabi, Ben Zhou
TL;DR
RELATE-Sim provides a theory-grounded, interactive simulator to forecast long-term romantic relationship dynamics by modeling dyadic behavior at consequential turning points. It combines two persona-aligned LLM agents with a Scene Master to generate a narrative of 3–4 auditable options per scene and infers interpretable states and commitment trajectories grounded in established relationship theory. Empirical evaluation on longitudinal data shows simulation-informed predictions outperform a personas-only baseline and reveal actionable markers, such as repair attempts and clarity shifts, that explain diverging trajectories. The framework shifts focus from matchmaking to maintenance, offering a transparent, extensible platform for understanding, forecasting, and potentially guiding long-term relationship dynamics in research and design contexts.
Abstract
Most dating technologies optimize for getting together, not staying together. We present RELATE-Sim, a theory-grounded simulator that models how couples behave at consequential turning points-exclusivity talks, conflict-and-repair episodes, relocations-rather than static traits. Two persona-aligned LLM agents (one per partner) interact under a centralized Scene Master that frames each turning point as a compact set of realistic options, advances the narrative, and infers interpretable state changes and an auditable commitment estimate after each scene. On a longitudinal dataset of 71 couples with two-year follow-ups, simulation-aware predictions outperform a personas-only baseline while surfacing actionable markers (e.g., repair attempts acknowledged, clarity shifts) that explain why trajectories diverge. RELATE-Sim pushes the relationship research's focus from matchmaking to maintenance, providing a transparent, extensible platform for understanding and forecasting long-term relationship dynamics.