{"ID":2852219,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.18551","arxiv_id":"2510.18551","title":"SOCIA-Nabla: Textual Gradient Meets Multi-Agent Orchestration for Automated Simulator Generation","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 -\u003e execution -\u003e evaluation -\u003e 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.","short_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 -\u003e execution -\u003e e...","url_abs":"https://arxiv.org/abs/2510.18551","url_pdf":"https://arxiv.org/pdf/2510.18551v2","authors":"[\"Yuncheng Hua\",\"Sion Weatherhead\",\"Mehdi Jafari\",\"Hao Xue\",\"Flora D. Salim\"]","published":"2025-10-21T12:00:00Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Large Language Model\"]","has_code":false}