Improving Large Vision-Language Models' Understanding for Flow Field Data

TL;DR

This paper introduces FieldLVLM, a framework that enhances large vision-language models' understanding of complex scientific field data by extracting key features and optimizing model tuning, leading to improved performance on scientific datasets.

Contribution

The paper presents a novel framework combining field-aware language generation and data compression for tuning LVLMs on scientific data, addressing a gap in applying these models to natural science domains.

Findings

FieldLVLM outperforms existing methods on scientific field data benchmarks.

The approach effectively extracts and encodes key physical features from complex data.

Experimental results demonstrate improved understanding of flow field data by LVLMs.

Abstract

Large Vision-Language Models (LVLMs) have shown impressive capabilities across a range of tasks that integrate visual and textual understanding, such as image captioning and visual question answering. These models are trained on large-scale image and video datasets paired with text, enabling them to bridge visual perception and natural language processing. However, their application to scientific domains, especially in interpreting complex field data commonly used in the natural sciences, remains underexplored. In this work, we introduce FieldLVLM, a novel framework designed to improve large vision-language models' understanding of field data. FieldLVLM consists of two main components: a field-aware language generation strategy and a data-compressed multimodal model tuning. The field-aware language generation strategy leverages a special-purpose machine learning pipeline to extract key physical features from field data, such as flow classification, Reynolds number, and vortex patterns. This information is then converted into structured textual descriptions that serve as a dataset. The data-compressed multimodal model tuning focuses on LVLMs with these generated datasets, using a data compression strategy to reduce the complexity of field inputs and retain only the most informative values. This ensures compatibility with the models language decoder and guides its learning more effectively. Experimental results on newly proposed benchmark datasets demonstrate that FieldLVLM significantly outperforms existing methods in tasks involving scientific field data. Our findings suggest that this approach opens up new possibilities for applying large vision-language models to scientific research, helping bridge the gap between large models and domain-specific discovery.