Supplementary Material for: Quantitative Evaluation and Domain Adaptation of Vision–Language Models for Mixed-Reality Interpretation of Indoor Environmental Computational Fluid Dynamics Visualizations

This dataset contains the supplymentary materials for the manuscripts submitted to MDPI Technologies 1. Data Folder This data folder contains the following subfolders: create_dataset : Scripts used for creating dataset are included here. dataset : The dataset constructed in this research is included here. Images : MR and colorbar images used in the dataset and background images for MR images are included here. inference : Scripts for inference are included in this folder. qwen2.5-vl-7B-splitdata: VLMs fine-tuned on the dataset (epoch-5 , epoch-10) are included here. results : Inference results of each model (the baseline model, epoch-5, epoch-10) are included here. temperature_csv : Temperature data at a specified cross-section from the CFD analysis results used to create the dataset is included here. train : Scripts for fine-tuning are included in this folder. All scripts in these subfolder must be run from the parent directory. 2. License Creative Commons Attribution 4.0 International (CC-BY 4.0) 3. Funding This research was funded by JSPS KAKENHI, grant number 23K11724.