mlsanityData to 14 Oct 2026

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

arXiv 2505.12006 ↗First posted 17 May 2025cs.AIcited by 1

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. We will release the code soon.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 30 of its 77 references that are in the map. First posted on arXiv 17 May 2025.

Builds on

30 of 77 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024RGD: Multi-LLM Based Agent Debugger via Refinement and Generation Guidance2 Oct
  2. 2024AutoGenesisAgent: Self-Generating Multi-Agent Systems for Complex Tasks25 Apr
  3. 2023ChatDev: Communicative Agents for Software Development16 Jul

Other references · 27

  1. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  2. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
  3. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Societycited by 1,395
  4. 2023Automatic Prompt Optimization with "Gradient Descent" and Beam Searchcited by 653
  5. 2024Hallucination is Inevitable: An Innate Limitation of Large Language Modelscited by 332
  6. 2014Caffe: Convolutional Architecture for Fast Feature Embeddingcited by 288
  7. 2024TextGrad: Automatic "Differentiation" via Textcited by 273
  8. 2023AutoAgents: A Framework for Automatic Agent Generationcited by 216
  9. 2024AgentSquare: Automatic LLM Agent Search in Modular Design Spacecited by 155
  10. 2023RecMind: Large Language Model Powered Agent For Recommendationcited by 129
  11. 2024Beyond Functional Correctness: Exploring Hallucinations in LLM-Generated Codecited by 107
  12. 2024The Butterfly Effect of Altering Prompts: How Small Changes and Jailbreaks Affect Large Language Model Performancecited by 100
15 more
  1. 2024Large Language Models as Urban Residents: An LLM Agent Framework for Personal Mobility Generationcited by 100
  2. 2024CodeR: Issue Resolving with Multi-Agent and Task Graphscited by 91
  3. 2024Automated Unit Test Improvement using Large Language Models at Metacited by 75
  4. 2015Fast low-rank estimation by projected gradient descent: General statistical and algorithmic guaranteescited by 70
  5. 2018How Many Random Seeds? Statistical Power Analysis in Deep Reinforcement Learning Experimentscited by 60
  6. 2023RCOT: Detecting and Rectifying Factual Inconsistency in Reasoning by Reversing Chain-of-Thoughtcited by 47
  7. 2024LLM Harmony: Multi-Agent Communication for Problem Solvingcited by 35
  8. 2024OpenCity: A Scalable Platform to Simulate Urban Activities with Massive LLM Agentscited by 35
  9. 2024Human-in-the-loop or AI-in-the-loop? Automate or Collaborate?cited by 23
  10. 2025AgentSociety Challenge: Designing LLM Agents for User Modeling and Recommendation on Web Platformscited by 12
  11. 2023Calibrating Neural Simulation-Based Inference with Differentiable Coverage Probabilitycited by 11
  12. 2020Universal Average-Case Optimality of Polyak Momentumcited by 9
  13. 2022BESSIE: A Behavior and Epidemic Simulator for Use With Synthetic Populationscited by 4
  14. 2020Human-in-the-Loop Design Cycles -- A Process Framework that Integrates Design Sprints, Agile Processes, and Machine Learning with Humanscited by 3
  15. 2020Joint Modelling of Cyber Activities and Physical Context to Improve Prediction of Visitor Behaviorscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026LangMARL: Natural Language Multi-Agent Reinforcement Learningcited by 4

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. 2026SOCIA-EVO: Automated Simulator Construction via Dual-Anchored Bi-Level Optimization
  7. 2026TRIM: Reducing AI-Generated CodeSlop via Agent Trajectory Minimization
  8. 2025Enhancing LLM Code Generation: A Systematic Evaluation of Multi-Agent Collaboration and Runtime Debugging for Improved Accuracy, Reliability, and Latency