SOCIA-Nabla: Textual Gradient Meets Multi-Agent Orchestration for Automated Simulator Generation

TL;DR

SOCIA-Nabla introduces an innovative framework that combines textual gradients with multi-agent orchestration to automate simulator generation, achieving state-of-the-art accuracy across diverse CPS tasks with minimal human effort.

Contribution

The paper presents SOCIA-Nabla, a novel end-to-end agentic framework that integrates LLM-driven agents and a loss-driven workflow for automated, scalable simulator code generation.

Findings

Achieves state-of-the-art accuracy on three CPS tasks.

Unifies multi-agent orchestration with loss-aligned optimization.

Converts brittle prompt pipelines into reproducible, constraint-aware code.

Abstract

In this paper, we present SOCIA-Nabla, an end-to-end, agentic framework that treats simulator construction asinstance optimization over code within a textual computation graph. Specialized LLM-driven agents are embedded as graph nodes, and a workflow manager executes a loss-driven loop: code synthesis -> execution -> evaluation -> code repair. The optimizer performs Textual-Gradient Descent (TGD), while human-in-the-loop interaction is reserved for task-spec confirmation, minimizing expert effort and keeping the code itself as the trainable object. Across three CPS tasks, i.e., User Modeling, Mask Adoption, and Personal Mobility, SOCIA-Nabla attains state-of-the-art overall accuracy. By unifying multi-agent orchestration with a loss-aligned optimization view, SOCIA-Nabla converts brittle prompt pipelines into reproducible, constraint-aware simulator code generation that scales across domains and simulation granularities. This work is under review, and we will release the code soon.