Cyclo-ETR-CBM: VLM-VQA-Guided Concept Bottleneck Learning for Interpretable Tropical Cyclone Intensity Estimation
Published: 12 July 2026| Version 1 | DOI: 10.17632/5dgxvr68sz.1
Contributors:
Abrar Faiaz, Mirza Rajit Raihan, Md. Mafizur Rahman, Md. Bazlur Rashid, S.M. Quamrul HasanDescription
This repository contains the data files prepared for the manuscript: Cyclo-ETR-CBM: VLM-VQA-Guided Concept Bottleneck Learning for Interpretable Tropical Cyclone Intensity Estimation The files are provided to support transparency and reproducibility of the reported experiments. The dataset includes the curated cyclone image and label files used for model development, along with the VLM-derived semantic feature files generated for the storm-wise train, validation, and test partitions.
Files
Categories
Interpretable Machine Learning