Toward Reliable Scientific Visualization Pipeline Construction with Structure-Aware Retrieval-Augmented LLMs

TL;DR

This paper presents a structure-aware retrieval-augmented approach to improve the reliability of LLM-generated scientific visualization pipelines, focusing on web-based vtk.js, by providing contextual guidance and reducing manual corrections.

Contribution

It introduces a retrieval-augmented workflow that supplies domain-specific code examples to enhance pipeline correctness and reduce manual correction efforts in scientific visualization.

Findings

Structured context improves pipeline executability.

Reduced manual correction cost with domain-specific guidance.

Enhanced human-in-the-loop inspection for pipeline validation.

Abstract

Scientific visualization pipelines encode domain-specific procedural knowledge with strict execution dependencies, making their construction sensitive to missing stages, incorrect operator usage, or improper ordering. Thus, generating executable scientific visualization pipelines from natural-language descriptions remains challenging for large language models, particularly in web-based environments where visualization authoring relies on explicit code-level pipeline assembly. In this work, we investigate the reliability of LLM-based scientific visualization pipeline generation, focusing on vtk.js as a representative web-based visualization library. We propose a structure-aware retrieval-augmented generation workflow that provides pipeline-aligned vtk.js code examples as contextual guidance, supporting correct module selection, parameter configuration, and execution order. We evaluate the proposed workflow across multiple multi-stage scientific visualization tasks and LLMs, measuring reliability in terms of pipeline executability and human correction effort. To this end, we introduce correction cost as metric for the amount of manual intervention required to obtain a valid pipeline. Our results show that structured, domain-specific context substantially improves pipeline executability and reduces correction cost. We additionally provide an interactive analysis interface to support human-in-the-loop inspection and systematic evaluation of generated visualization pipelines.