Toward Causal-Visual Programming: Enhancing Agentic Reasoning in Low-Code Environments

TL;DR

The paper addresses hallucinations and brittle reasoning in LLM-powered agents operating in low-code environments by introducing Causal-Visual Programming (CVP), which makes causal structure a first-class citizen through a DAG-based world model $G=(V,E)$. CVP provides a visual interface for constructing this causal graph and enforces causal constraints during inference (causal anchoring), restricting the agent to the causal parents of each target module. The authors formalize the framework, implement the constraint mechanism (via methods like feature selection and dynamic planning filtering), and validate the approach with a synthetic distribution-shift experiment, showing that a causal-anchored model maintains high accuracy under shift while an associative model does not. This demonstrates CVP’s potential to enhance robustness, interpretability, and trust in AI agents for dynamic, high-stakes settings by aligning agent reasoning with human causal knowledge and explicit world models.

Abstract

Large language model (LLM) agents are increasingly capable of orchestrating complex tasks in low-code environments. However, these agents often exhibit hallucinations and logical inconsistencies because their inherent reasoning mechanisms rely on probabilistic associations rather than genuine causal understanding. This paper introduces a new programming paradigm: Causal-Visual Programming (CVP), designed to address this fundamental issue by explicitly introducing causal structures into the workflow design. CVP allows users to define a simple "world model" for workflow modules through an intuitive low-code interface, effectively creating a Directed Acyclic Graph (DAG) that explicitly defines the causal relationships between modules. This causal graph acts as a crucial constraint during the agent's reasoning process, anchoring its decisions to a user-defined causal structure and significantly reducing logical errors and hallucinations by preventing reliance on spurious correlations. To validate the effectiveness of CVP, we designed a synthetic experiment that simulates a common real-world problem: a distribution shift between the training and test environments. Our results show that a causally anchored model maintained stable accuracy in the face of this shift, whereas a purely associative baseline model that relied on probabilistic correlations experienced a significant performance drop. The primary contributions of this study are: a formal definition of causal structures for workflow modules; the proposal and implementation of a CVP framework that anchors agent reasoning to a user-defined causal graph; and empirical evidence demonstrating the framework's effectiveness in enhancing agent robustness and reducing errors caused by causal confusion in dynamic environments. CVP offers a viable path toward building more interpretable, reliable, and trustworthy AI agents.