# Leveraging Large Language Models for Automated Causal Loop Diagram Generation: Enhancing System Dynamics Modeling through Curated Prompting Techniques

arXiv: https://arxiv.org/abs/2503.21798 · HTML: https://mlsanity.com/paper/2503.21798/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 22 references that are in the map.
- First posted on arXiv 23 Mar 2025.

## Builds on (most cited first, 6 of 6)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing](https://mlsanity.com/paper/2107.13586/) (2021)
- [Can Foundation Models Talk Causality?](https://mlsanity.com/paper/2206.10591/) (2022)
- [Towards Understanding How Machines Can Learn Causal Overhypotheses](https://mlsanity.com/paper/2206.08353/) (2022)
- [From Text to Map: A System Dynamics Bot for Constructing Causal Loop Diagrams](https://mlsanity.com/paper/2402.11400/) (2024)

## Built on by (most cited first, 1 of 1)

- [CEDAR: Agent-Orchestrated Tree Search for Goal-Directed Optimization of Complex Systems](https://mlsanity.com/paper/2608.06871/) (2026)

## Abstract (arXiv)

Transforming a dynamic hypothesis into a causal loop diagram (CLD) is crucial for System Dynamics Modelling. Extracting key variables and causal relationships from text to build a CLD is often challenging and time-consuming for novice modelers, limiting SD tool adoption. This paper introduces and tests a method for automating the translation of dynamic hypotheses into CLDs using large language models (LLMs) with curated prompting techniques. We first describe how LLMs work and how they can make the inferences needed to build CLDs using a standard digraph structure. Next, we develop a set of simple dynamic hypotheses and corresponding CLDs from leading SD textbooks. We then compare the four different combinations of prompting techniques, evaluating their performance against CLDs labeled by expert modelers. Results show that for simple model structures and using curated prompting techniques, LLMs can generate CLDs of a similar quality to expert-built ones, accelerating CLD creation.