基于知识增强和对抗训练的立场检测方法

黄莉媛; 张瑾; 靳小龙; 徐辉; 郭嘉丰

doi:10.11959/j.issn.2096-0271.25261

您当前的位置：

首页 >

文章列表页 >

基于知识增强和对抗训练的立场检测方法

更新时间：2026-02-27

- 基于知识增强和对抗训练的立场检测方法
- Stance detection method based on knowledge augmentation and adversarial training
- 大数据 2026年页码：1-13
- 作者机构：
  
  1.中国科学院计算技术研究所网络数据科学与技术重点实验室，北京 100190
  2.智能算法安全全国重点实验室，北京 100190
  3.中国科学院大学计算机科学与技术学院，北京 101408
- 作者简介：
  
  [ "黄莉媛（2003- ），女，中国科学院计算技术研究所硕士生，主要研究方向为自然语言处理。" ]
  [ "张瑾（1978- ），男，博士，中国科学院计算技术研究所高级工程师，主要研究领域为自然语言处理、大数据处理、信息检索等。 E-mail： jinzhang@ict.ac.cn" ]
  [ "靳小龙（1976- ），男，博士，中国科学院计算技术研究所研究员，主要研究方向为知识图谱、知识计算、大数据知识工程等。" ]
  [ "徐辉（1989- ），男，硕士，中国科学院计算技术研究所工程师，主要研究方向为自然语言处理。" ]
  [ "郭嘉丰（1980- ），男，博士，中国科学院计算技术研究所研究员，现任中国科学院计算技术研究所网络数据科学与技术重点实验室主任，大数据分析系统国家工程研究中心常务副主任，主要研究方向为信息检索、自然语言理解、大数据分析系统等。" ]
- 基金信息：
- DOI：10.11959/j.issn.2096-0271.25261
  中图分类号： TP391.1
- 收稿：2025-11-30，
  
  网络首发：2026-01-26，
- 稿件说明：
移动端阅览
黄莉媛,张瑾,靳小龙等.基于知识增强和对抗训练的立场检测方法[J].大数据,

HUANG Liyuan,ZHANG Jin,JIN Xiaolong,et al.Stance detection method based on knowledge augmentation and adversarial training[J].BIG DATA RESEARCH,
黄莉媛,张瑾,靳小龙等.基于知识增强和对抗训练的立场检测方法[J].大数据, DOI：10.11959/j.issn.2096-0271.25261.

HUANG Liyuan,ZHANG Jin,JIN Xiaolong,et al.Stance detection method based on knowledge augmentation and adversarial training[J].BIG DATA RESEARCH, DOI：10.11959/j.issn.2096-0271.25261.

摘要

立场检测是自然语言处理领域的一个重要研究方向，旨在判断作者对特定目标所持有的支持、反对或中立态度。针对社交媒体场景中文本语言复杂、领域知识匮乏和模型泛化能力弱的问题，提出了一种基于知识增强和对抗训练的文本立场检测方法KABERT。方法结合生成式与判别式模型的优势，先使用生成式模型从关键词、隐含情感和修辞手法角度提取文本与目标之间的深层语义关系，生成隐含知识；再使用判别式模型作为主干分类网络进行有监督微调，并引入快速梯度法进行对抗训练。在两个标准立场检测数据集SEM16和P-Stance上的实验结果显示，所提方法的Macro-F1分别为68.85%和79.24%，较目前主流方法均有不同程度的提升，证明了所提方法的有效性。

Abstract

Stance detection is a crucial research direction in natural language processing

aiming to determine the author's supportive

opposing

or neutral attitude towards a specific target. To address challenges such as complex linguistic expressions

insufficient domain-specific knowledge

and weak model generalization in social media contexts

this paper proposed a stance detection method named KABERT

based on knowledge augmentation and adversarial training. The method integrated the advantages of generative and discriminative models. A generative model was first used to extract deep semantic relationships between text and the target from the perspectives of keywords

implicit sentiment

and rhetorical devices

generating implicit knowledge. Then

a discriminative model was used as the classification backbone network for supervised fine-tuning

and a fast gradient method was introduced for adversarial training. Experiments were conducted on two standard stance detection datasets

SEM16 and P-Stance. Results show that KABERT achieves Macro-F1 scores of 68.85% and 79.24% respectively

outperforming current mainstream approaches by varying margins

demonstrating the effectiveness of the proposed approach.

关键词

Keywords

references

Zhang B , Dai G , Niu F , et al . A Survey of Stance Detection on Social Media: New Directions and Perspectives [J ] . arXiv preprint arXiv: 2409.15690 , 2024 .

Du J , Xu R , He Y , et al . Stance classification with target specific neural attention networks [C ] // International Joint Conferences on Artificial Intelligence , 2017 . 3988 - 3994 .

白静 , 李霏 , 姬东鸿 . 基于注意力的BiLSTM-CNN中文微博立场检测模型 [J ] . 计算机应用与软件 , 2018 , 35 ( 3 ): 266 - 274 .

Bai J , Li F , Ji D H . Attention Based BiLSTM-CNN Chinese Microblogging Position Detection Model . Computer Applications and Software , 2018 , 35 ( 3 ): 266 - 274 .

杨顺成 , 李彦 , 赵其峰 . 基于GCN和Bi-LSTM的微博立场检测方法 [J ] . 重庆理工大学学报(自然科学) , 2020 , 34 ( 6 ): 167 - 173 .

Yang S C , Li Y , Zhao Q F . Stance Detection Method of Chinese Micro-Blog Based on GCN and Bi-LSTM . Journal of Chongqing University of Technology(Natural Science) , 2020 , 34 ( 6 ): 167 - 173 .

Radford A , Narasimhan K , Salimans T , et al . Improving language understanding by generative pre-training [J ] . 2018 .

Devlin , Jacob , et al . BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding [C ] // Proceedings of NAACL-HLT . 2019 : 4171 - 4186 .

Nguyen D Q , Vu T , Nguyen A T . Bertweet: A pre-trained language model for English Tweets [J ] . arXiv preprint arXiv: 2005.10200 , 2020 .

Li Y J , T S , A S , et al . 2021 . P-Stance: A Large Dataset for Stance Detection in Political Domain . In Findings of the Association for Computational Linguistics : ACL-IJCNLP 2021 , pages 2355 – 2365 .

Huang H , Zhang B W , Li Y Y , et al . 2023 . Knowledge-enhanced prompt-tuning for stance detection. ACM Trans. Asian Low Resour. Lang. Inf . Process. , 22 ( 6 ): 159 : 1 – 159 : 20 .

Li A , Liang B , Zhao J Q , et al . 2023 . Stance Detection on Social Media with Background Knowledge . In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 15703 – 15717 , Singapore. Association for Computational Linguistics.

Goodfellow I J , Shlens J , Szegedy C . Explaining and harnessing adversarial examples [J ] . arXiv preprint arXiv: 1412.6572 , 2014 .

Miyato T , Dai A M , Goodfellow I . Adversarial training methods for semi-supervised text classification [J ] . arXiv preprint arXiv: 1605.07725 , 2016 .

Madry A , Makelov A , Schmidt L , et al . Towards deep learning models resistant to adversarial attacks [J ] . arXiv preprint arXiv: 1706.06083 , 2017 .

Zhu D , Lin W , Zhang Y , et al . At-bert: Adversarial training bert for acronym identification winning solution for sdu@ aaai-21 [J ] . arXiv preprint arXiv: 2101.03700 , 2021 .

Zhang Z , Li Y , Zhang J , et al . Llm-driven knowledge injection advances zero-shot and cross-target stance detection [C ] // Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers) . 2024 : 371 - 378 .

Vaswani A , Shazeer N , Parmar N , et al . Attention is all you need [J ] . Advances in neural information processing systems , 2017 , 30 .

Karimi A , Rossi L , Prati A . Adversarial training for aspect-based sentiment analysis with bert [C ] // 2020 25th International conference on pattern recognition (ICPR) . IEEE , 2021 : 8797 - 8803 .

Saif M , Svetlana K , Parinaz S , et al . Semeval-2016 Task 6: Detecting Stance in Tweets . In Proceedings of the International Workshop on Semantic Evaluation (SemEval ’ 16 ). June 2016. San Diego, California .

Graves A , Schmidhuber J . Framewise phoneme classification with bidirectional LSTM and other neural network architectures [J ] . Neural networks , 2005 , 18 ( 5-6 ): 602 - 610 .

Augenstein I , RocktÄSchel T , Vlachos A , et al . Stance Detection with Bidirectional Conditional Encoding [C ] . Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing , 2016 , pp. 876 – 885 .

Kim Y . Convolutional neural networks for sentence classification [J ] . Empirical Methods in Natural Language Processing , 2014 : 1746 - 1751 .

Liu Y H , Ott M , Goyal N , et al . RoBERTA: a robustly optimized BERT pretraining approach [J ] . arXiv preprint arXiv: 1907.11692 , 2019 .

Yao Z , Yang W , Wei F . Enhancing zero-shot stance detection with contrastive and prompt learning [J ] . Entropy , 2024 , 26 ( 4 ): 325 .

Nguyen Q M , Kim T . Is External Information Useful for Stance Detection with LLMs? [J ] . arXiv preprint ， 2019 ，arXiv: 2507.01543 , 2025.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

杉杉：面向高吞吐低延迟服务的计算机公共课问答系统

基于知识增强大语言模型的历史学科试题生成系统

基于重排序和后检索反思的教育大模型问答增强方法

面向教育场景的视觉大模型优化与应用

面向法律领域的大模型微调与应用