2026-10-13T22:13:11Z https://www.scidb.cn/oai

10.57760/sciencedb.j00001.01727 2026-08-21T11:27:29Z

2026-08-21 XBMU-TibSum: A Tibetan Text Summarization Dataset for Large Language Model Fine-Tuning doi:10.57760/sciencedb.j00001.01727 en A Tibetan long-text summarization dataset constructed for large language model (LLM) fine-tuning. It contains 9,979 pairs of structured instruction fine-tuning samples. The dataset utilizes LLMs to reconstruct high-information-density summary labels and extracts core entities to serve as semantic anchors. Large language models; Long-textsummarization; Tibetan; Instructionfine-tuning; Keyword planning dan rui lei He Xiangzhen Meng Xianghe Ouyang Xu Huangzijian Shi Shibiao Zhu Haotian Hao Jingbin Wang Xiang Jiang Wenhao Du Tingting PUBLIC https://creativecommons.org/licenses/by-nc/4.0/ dataset Science Data Bank