重庆邮电大学学报（自然科学版）2025，Vol.37Issue(5)：688-695,8.DOI:10.3979/j.issn.1673-825X.202409020228
面向生成式文本摘要模型的内在幻觉优化方法
An intrinsic hallucination optimization method for generative text summarization models
摘要
Abstract
Generative text summarization models can produce novel expressions in summaries,but even the most advanced models may generate content that contradicts the source text or lacks factual verifiability—a phenomenon known as halluci-nation.To address this issue,this paper proposes an intrinsic hallucination optimization method to improve the summariza-tion generation process.The proposed approach mitigates hallucinations from three perspectives:data-level optimization,model training-level optimization,and summary generation strategy-level optimization.Experiments conducted on two benchmark datasets demonstrate the superior performance of the proposed method.Compared with baseline models,the pro-posed approach achieves an average improvement of 8.58％in R-1 score on the CNNDM dataset and 7.26％on the XSUM dataset.The results indicate that the method not only enhances summary quality but also effectively reduces hallucinations,providing a valuable reference for the practical deployment of generative text summarization models.关键词
生成式文本摘要/内在幻觉/候选摘要/大语言模型Key words
generative text summarization/intrinsic hallucination/candidate summaries/large language model分类
计算机与自动化引用本文复制引用
李能,于成成,刘群..面向生成式文本摘要模型的内在幻觉优化方法[J].重庆邮电大学学报（自然科学版）,2025,37(5):688-695,8.基金项目
国家自然科学基金重点项目(61936001) （61936001）
重庆市教委重点合作项目(HZ2021008)Key Project of National Natural Science Foundation of China(61936001) （HZ2021008）
Key Cooperation Project of Chongqing Municipal Education Commission(HZ2021008) （HZ2021008）
