计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (6A): 240700131-5.doi: 10.11896/jsjkx.240700131

• 大语言模型技术及应用 • 上一篇    下一篇

大模型识别谣言不同来源效能研究

何静1, 陈逸然2   

  1. 1 北京航空航天大学人文与社会科学高等研究院 北京 100191
    2 北京航空航天大学人工智能学院 北京 100191
  • 出版日期:2025-06-16 发布日期:2025-06-12
  • 通讯作者: 何静(bhhejing@buaa.edu.cn)
  • 基金资助:
    北京市社会科学基金(23XCC020);北京市教育科学“十四五”规划课题(CGCA23128)

Study on Efficiency of Large Model in Recognizing Rumors from Different Sources

HE Jing1, CHEN Yiran2   

  1. 1 Institute for Advanced Studied in Humanities and Social Sciences,Beihang University,Beijing 100191,China
    2 School of Artificial Intelligence,Beihang University,Beijing 100191,China
  • Online:2025-06-16 Published:2025-06-12
  • About author:HE Jing,born in 1989,Ph.D,assistant professor.Her main research interests include AI and big data.
  • Supported by:
    Beijing Social Science Fund(23XCC020) and Beijing Education Science 14th Five Year Plan Project(CGCA23128).

PDF (PC)

59

摘要： 针对网络谣言识别面临的新挑战,探索大模型在识别谣言不同来源中的效能。研究构建国内外人为谣言与AI谣言数据集,据此在零样本设置情况下,对4种大模型的谣言来源辨识能力进行测试。研究发现,单一大模型识别谣言的精确度较低,存在明显错误倾向。为提高识别性能,采用预训练、微调和集成学习等方法,使得大模型性能得到显著提升。进一步,提出基于模型碰撞的集成学习方法,利用多模型反馈改善谣言来源识别效能。实验结果显示,集成学习框架能够整合各模型优势,显著提高识别准确性。通过实证研究验证了大型语言模型在谣言识别中的潜力和改进方向,有助于应对当前复杂的网络谣言环境,维护网络空间的清朗。

关键词: AIGC, 大模型, 谣言识别, 谣言治理

Abstract: This study aims to address the new challenges faced by online rumor recognition and explore the effectiveness of large models in recognizing different sources of rumors.Constructing domestic and foreign rumor and AI rumor datasets,and testing the rumor source identification ability of four large models under zero sample settings.Research has found that a single large model has low accuracy in identifying rumors and has a clear tendency towards errors.To improve recognition performance,me-thods such as pre-training,fine-tuning,and ensemble learning are adopted to significantly enhance the performance of the large model.Furthermore,a model collision based ensemble learning method is proposed to improve the effectiveness of rumor source recognition by utilizing multi model feedback.Experimental results show that the ensemble learning framework can integrate the advantages of various models and significantly improve recognition accuracy.This study verifies the potential and improvement direction of large-scale language models in rumor recognition through empirical research,which helps to cope with the current complex online rumor environment and maintain the clarity of cyberspace.

Key words: AIGC, Large model, Rumor recognition, Rumor control

中图分类号: 

  • G206
[1]YAN Y X,LI B,FENG J Y,et al.Research on the impact of trends related to ChatGPT[J].Procedia Computer Science,2023,221:1281-1294.
[2]HUTSON M.Robo-writers:the rise and risks of language-ge-nerating AI[J].Nature,2021,591(7848):22-25.
[3]LI H,MOON J,PURKAYASTHA S,et al.Ethics of large language models in medicine and medical research[J].The Lancet Digital Health,2023,5(6):333-335.
[4]LIU Y,SHEN H,SHI L.A review of rumor detection tech-niques in social networks[J].Journal of Intelligent & Fuzzy Systems,2023,44(3):3561-3578.
[5]KWON S,CHA M,JUNG K,et al.Prominent Features of Rumor Propagation in Online Social Media[C]//Proceedings of the 13th International Conference on Data Mining.New York:IEEE Press,2013:1103-1108.
[6]YANG C,ZHANG P,QIAO W,et al.Rumor Detection on So-cial Media with Crowd Intelligence and ChatGPT-Assisted Networks[C]//Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing.2023:570-5717.
[7]VARLAMIS I,MICHAIL D,GLYKOU F,et al.A survey onthe use of graph convolutional networks for combating fake news[J].Future Interne,2022,14:70.
[8]SHAHID W,LI Y R,STAPLES D,et al.Are You a Cyborg,Bot or Human?－A Survey on Detecting Fake News Spreaders[J].IEEE Access,2022,10:27069-27083.
[9]D’ULIZIA A,CASCHERA C M,FERRI F,et al.Fake newsdetection:a survey of evaluation datasets[J].PeerJ Computer Science,2021,7:518.
[10]ALSAIF H F,ALDOSSARI H D.Review of stance detection for rumor verification in social media[J].Engineering Applications of Artificial Intelligence,2023,119:105801.
[11]HAMDA S,IBRAHIM B,YAHYA S.Adapting Pre-trainedLanguage Models to Rumor Detection on Twitter[J].Journal of Universal Computer Science,2021,27(10):1128-1148.
[12]CHEN C Y,SHU K.2023.Combating misinformation in the age of LLMs:Opportunities and challenges[J].AI Magazine,2024,45(3):1-15.
[13]LIU Q,TAO X,WU J F,et al.Can Large Language Models Detect Rumors on Social Media?[J].arXiv:2402.03916,2024.
[14]YAN Y Q,ZHENG P,WANG Y J.Enhancing large language model capabilities for rumor detection with Knowledge-Powered Prompting[J].Engineering Applications of Artificial Intelligence,2024,133(C):108259.
[15]BANG Y J,CAHYAWIJAYA S,LEE N,et al.A Multitask,Multilingual,Multimodal Evaluation of ChatGPT on Reasoning,Hallucination,and Interactivity[J].arXiv:2302.04023,2023.
[16]ALI H,QADIR J,SHAH Z,et al.ChatGPT and large language models(LLMs) in healthcare:Opportunities and risks[C]//Proceedings of the 2023 IEEE International Conference on Artificial Intelligence,Blockchain,and Internet of Things.New York:IEEE Press,2023:1-4.
[17]ZHANG Y Y,YUAN J J.A review of rumor detection research based on social media[J].Data Communication,2024(1):28-33.
[18]ZAEEM R N,LI C J,BARBER K S,et al.On sentiment of online fake news[C]//Proceedings of 2020 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining.New York:IEEE Press,2020:760-767.
[19]ZHANG N,SONG J H,CHEN K,et al.Emotional contagion in the propagation of online rumors[J].Issues in Information Systems,2022,23(2):1-19.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发