Bingxin Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Bingxin Li has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Ying Guo
North China University of Technology, Beijing, China
2 shared publications - Chong Di
Shandong Artificial Intelligence Institute, Qilu University of Technology (Shandong Academy of Sciences), Jinan, China
1 shared publication - Gaolei Li
Shanghai Jiao Tong University, Shanghai, China
1 shared publication - Hong Ge
North China University of Technology
1 shared publication - Jie Liu
School of Artificial Intelligence and Computer Science, North China University of Technology, Beijing 100144, China
1 shared publication - Kexin Zhen
North China University of Technology, Beijing, P. R. China
1 shared publication
Latest Publications
- 13 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 6 (7) , 1787-1796
Consistency-Heterogenity Balanced Fake News Detection via Cross-Modal Matching
Generating synthetic content through Generative AI (GAI) presents considerable hurdles for current fake news detection methodologies. Many existing detection approaches concentrate on feature-based multi-modal fusion, neglecting semantic relationships such as correlations and diversities. In this study, we introduce an innovative cross-modal matching-driven approach to reconcile semantic relevance (text-image consistency) and semantic gap (text-image heterogeneity) in multimodal fake news detection. Unlike the conventional paradigm of multi-modal fusion followed by detection, our approach integrates textual modality, visual modality (images), and text embedded within images (auxiliary modality) to construct an end-to-end framework. This framework considers the relevance of contents across different modalities while simultaneously addressing the gap in structures, achieving a delicate balance between consistency and heterogeneity. Consistency is fostered by evaluating inter-modality correlation via pairwise-similarity scores, while heterogeneity is addressed by employing cross-attention mechanisms to account for inter-modality diversity. To achieve equilibrium between consistency and heterogeneity, we employ attention-guided enhanced modality interaction and similarity-based dynamic weight assignment to establish robust frameworks. Comparative experiments conducted on the Chinese Weibo dataset and the English Twitter dataset demonstrate the effectiveness of our approach, surpassing the state-of-the-art by 7% to 13%.
- 13 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 6 (7) , 1787-1796
Consistency-Heterogenity Balanced Fake News Detection via Cross-Modal Matching
Generating synthetic content through Generative AI (GAI) presents considerable hurdles for current fake news detection methodologies. Many existing detection approaches concentrate on feature-based multi-modal fusion, neglecting semantic relationships such as correlations and diversities. In this study, we introduce an innovative cross-modal matching-driven approach to reconcile semantic relevance (text-image consistency) and semantic gap (text-image heterogeneity) in multimodal fake news detection. Unlike the conventional paradigm of multi-modal fusion followed by detection, our approach integrates textual modality, visual modality (images), and text embedded within images (auxiliary modality) to construct an end-to-end framework. This framework considers the relevance of contents across different modalities while simultaneously addressing the gap in structures, achieving a delicate balance between consistency and heterogeneity. Consistency is fostered by evaluating inter-modality correlation via pairwise-similarity scores, while heterogeneity is addressed by employing cross-attention mechanisms to account for inter-modality diversity. To achieve equilibrium between consistency and heterogeneity, we employ attention-guided enhanced modality interaction and similarity-based dynamic weight assignment to establish robust frameworks. Comparative experiments conducted on the Chinese Weibo dataset and the English Twitter dataset demonstrate the effectiveness of our approach, surpassing the state-of-the-art by 7% to 13%.
- 22 September 2023
- Published by Springer Nature
- p. 65-76
An Auxiliary Modality Based Text-Image Matching Methodology for Fake News Detection
- 22 September 2023
- Published by Springer Nature
- p. 65-76
An Auxiliary Modality Based Text-Image Matching Methodology for Fake News Detection