mlsanityData to 7 Oct 2026

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

arXiv 2510.18551 ↗First posted 21 Oct 2025cs.AIcited by 0

On the map: LLM agents and tool use → Software engineering coding agents

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 33 of its 81 references that are in the map. First posted on arXiv 21 Oct 2025.

Builds on

33 of 81 references.

Closest ancestors · 5

Same direction, newest first

  1. 2024RGD: Multi-LLM Based Agent Debugger via Refinement and Generation Guidance2 Oct
  2. 2024From LLMs to LLM-based Agents for Software Engineering: A Survey of Current, Challenges and Future5 Aug
  3. 2024AutoGenesisAgent: Self-Generating Multi-Agent Systems for Complex Tasks25 Apr
  4. 2023AgentCoder: Multi-Agent-based Code Generation with Iterative Testing and Optimisation20 Dec
  5. 2023ChatDev: Communicative Agents for Software Development16 Jul

Other references · 28

  1. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  2. 2025The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Searchcited by 375
  3. 2024Hallucination is Inevitable: An Innate Limitation of Large Language Modelscited by 329
  4. 2024A Comprehensive Survey of Hallucination Mitigation Techniques in Large Language Modelscited by 266
  5. 2024TextGrad: Automatic "Differentiation" via Textcited by 250
  6. 2025AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Societycited by 180
  7. 2024AgentSquare: Automatic LLM Agent Search in Modular Design Spacecited by 141
  8. 2024OASIS: Open Agent Social Interaction Simulations with One Million Agentscited by 127
  9. 2024Beyond Functional Correctness: Exploring Hallucinations in LLM-Generated Codecited by 106
  10. 2024The Butterfly Effect of Altering Prompts: How Small Changes and Jailbreaks Affect Large Language Model Performancecited by 97
  11. 2024CodeR: Issue Resolving with Multi-Agent and Task Graphscited by 89
  12. 2024Automated Unit Test Improvement using Large Language Models at Metacited by 73
16 more
  1. 2015Fast low-rank estimation by projected gradient descent: General statistical and algorithmic guaranteescited by 68
  2. 2018How Many Random Seeds? Statistical Power Analysis in Deep Reinforcement Learning Experimentscited by 56
  3. 2025SocioVerse: A World Model for Social Simulation Powered by LLM Agents and A Pool of 10 Million Real-World Userscited by 50
  4. 2024GenSim: A General Social Simulation Platform with Large Language Model based Agentscited by 48
  5. 2023RCOT: Detecting and Rectifying Factual Inconsistency in Reasoning by Reversing Chain-of-Thoughtcited by 47
  6. 2024OpenCity: A Scalable Platform to Simulate Urban Activities with Massive LLM Agentscited by 35
  7. 2024LLM Harmony: Multi-Agent Communication for Problem Solvingcited by 34
  8. 2025Can LLM Agents Simulate Multi-Turn Human Behavior? Evidence from Real Online Customer Behavior Datacited by 27
  9. 2025TrajLLM: A Modular LLM-Enhanced Agent-Based Framework for Realistic Human Trajectory Simulationcited by 21
  10. 2025AgentSociety Challenge: Designing LLM Agents for User Modeling and Recommendation on Web Platformscited by 12
  11. 2025YuLan-OneSim: Towards the Next Generation of Social Simulator with Large Language Modelscited by 11
  12. 2025Foam-Agent: A Large Language Model-Based Multi-Agent Framework for Automating Computational Fluid Dynamics Workflowscited by 9
  13. 2024LLMAuditor: A Framework for Auditing Large Language Models Using Human-in-the-Loopcited by 8
  14. 2025G-Sim: Generative Simulations with Large Language Models and Gradient-Free Calibrationcited by 8
  15. 2022BESSIE: A Behavior and Epidemic Simulator for Use With Synthetic Populationscited by 3
  16. 2023Masking Behaviors in Epidemiological Networks with Cognitively-plausible Reinforcement Learningcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025SOCIA-nabla: Textual Gradient Meets Multi-Agent Orchestration for Automated Simulator Generation
  2. 2025CODESIM: Multi-Agent Code Generation and Problem Solving through Simulation-Driven Planning and Debugging
  3. 2024Self-Organized Agents: A LLM Multi-Agent Framework toward Ultra Large-Scale Code Generation and Optimization
  4. 2025SEW: Self-Evolving Agentic Workflows for Automated Code Generation
  5. 2024MapCoder: Multi-Agent Code Generation for Competitive Problem Solving
  6. 2025Enhancing LLM Code Generation: A Systematic Evaluation of Multi-Agent Collaboration and Runtime Debugging for Improved Accuracy, Reliability, and Latency
  7. 2026SOCIA-EVO: Automated Simulator Construction via Dual-Anchored Bi-Level Optimization
  8. 2026CAM: A Causality-based Analysis Framework for Multi-Agent Code Generation Systems