SOCIA-EVO: Automated Simulator Construction via Dual-Anchored Bi-Level Optimization

ACL ARR 2026 January Submission7518 Authors

Keywords: Automated Simulator Construction, Dual-Anchor, Bi-Level Optimization, Strategy Playbook, Distributional Fidelity

Abstract: Automated simulator construction requires distributional fidelity, distinguishing it from generic code generation. We identify two failure modes in long-horizon LLM agents: contextual drift and optimization instability arising from conflating structural and parametric errors. We propose SOCIA-EVO, a dual-anchored evolutionary framework. SOCIA-EVO introduces: (1) a static blueprint to enforce empirical constraints; (2) a bi-level optimization to decouple structural refinement from parameter calibration; and (3) a self-curating Strategy Playbook that manages remedial hypotheses via Bayesian-weighted retrieval. By falsifying ineffective strategies through execution feedback, SOCIA-EVO achieves robust convergence, generating simulators that are statistically consistent with observational data. SOCIA-EVO’s code and data are available here: https://anonymous.4open.science/r/SOCIA-F80E.

Paper Type: Long

Research Area: AI/LLM Agents

Research Area Keywords: AI / LLM Agents, Code Models

Contribution Types: NLP engineering experiment

Languages Studied: English

Submission Number: 7518