网络与信息安全学报2026，Vol.12Issue(2)：156-168,13.DOI:10.11959/j.issn.2096-109x.25249
DynaSynth:面向非结构化网络安全语料的大语言模型微调数据动态生成代理工具
DynaSynth:a large language model fine-tuning data dynamic synthesis agent tool for unstructured cybersecurity corpus
摘要
Abstract
With the rapid development of model training technologies and data scale,large language models have been able to handle various general tasks.However,in the vertical field of cybersecurity,the development of large language models is still constrained by data bottlenecks.To solve this problem,a data dynamic generation agent tool named DynaSynth for large language models based on unstructured text corpora was proposed.This tool first uses vector embeddings to assist the model in splitting the original text into blocks and converting them into retriev-able enhanced vector data;then,through audience and genre analysis of the documents,it guides the generation of data with diverse question-answer pair styles and universality.During this process,users can view the generated data in real time through a visual interface and optimize the generation strategy by leveraging prompt engineering to ensure that the generated content precisely meets the actual needs.Experimental results show that the fine-tuning data generated by DynaSynth can significantly improve the performance of large language models in vertical do-main tasks.This tool not only has an efficient data generation capability,but also effectively explores the diverse generation paths of data.关键词
大语言模型代理/生成数据/数据增强Key words
large language model agent/synthesis data/data augmentation分类
信息技术与安全科学引用本文复制引用
关永健,朴乘锴,王布宏,赵博夫,李思琦,赵正阳..DynaSynth:面向非结构化网络安全语料的大语言模型微调数据动态生成代理工具[J].网络与信息安全学报,2026,12(2):156-168,13.基金项目
国家自然科学基金资助项目(No.62472437) The National Natural Science Foundation of China(No.62472437) （No.62472437）
