Supporting material for the research titled "Advances in hydrological research in China over the past two decades: insights from advanced large language model and topic modeling"

This dataset is the supporting material for the research titled "Advances in hydrological research in China over the past two decades: insights from advanced large language model and topic modeling" (Hu et al., 2025). It consists of three components:1. Wizard-Vicuna-30B-Uncensored-GPTQ: This folder contains a large language model designed for intelligent parsing the hydrological information from a vast body of publications. The model is based on LLaMA 30B and fine-tuned using the Wizard-Vicuna dataset. To reduce memory requirements, GPTQ (4-bit quantization) was applied. The model can be used via the Hugging Face Transformers library. For more information about this model, please refer to Wizard-Vicuna-30B-Uncensored-GPTQ on Hugging Face (https://huggingface.co/TheBloke/Wizard-Vicuna-30B-Uncensored-GPTQ).2. parameters of LLM.xlsx: This file contains the parameters used for parsing hydrological information with the large language model, including quantization settings and inference parameters.3. parsing prompt.txt: This file contains the prompt used during information extraction. In this prompt, "TITLE" and "ABSTRACT" correspond to the title and abstract of a given publication, respectively.The publications used for data parsing can be accessed on Zenodo ( https://doi.org/10.5281/zenodo.11648428). The dataset was compiled from a search performed in October 2023 using the Web of Science (WoS). It covers information including title, abstract, journal name, author names, author affiliations, author keywords, keywords generated by WoS, and other related metadata.For more details on data acquisition and intelligent hydrological information parsing, please refer to Hu et al. (2025) and Miao et al. (2024).

References:1. Hu, J., Miao, C., Wu, Y., & Su, J. Advances in hydrological research in China over the past two decades: insights from advanced large language model and topic modeling. Fundamental Research. in review, 2025

2. Miao, C., Hu, J., Moradkhani, H., & Destouni, G. (2024). Hydrological Research Evolution: A Large Language Model-Based Analysis of 310,000 Studies Published Globally Between 1980 and 2023. Water Resources Research, 60(6), e2024WR038077. https://doi.org/10.1029/2024WR038077