Improving Entity Understanding for Vision-Language Pre-Training via Active Learning
Big Data and Cognitive Computing · 22.06.2026 · 10 (6) · 198
DOI: 10.3390/bdcc10060198 ↗Kurz gefasst
Fachartikel in der Zeitschrift Big Data and Cognitive Computing, online erschienen am 22.06.2026. Verfasst von Qunbo Wang und 8 weiteren Autor:innen. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shanghai Aerospace Science and Technology Innovation Foundation, Aviation Science Fund, Beijing Natural Science Foundation.
Abstract (Originalsprache)
Although many researchers use pre-trained models to better solve downstream tasks, further exploration of more effective pre-training methods remains necessary, especially for multi-modal pre-training where high-quality training data is more difficult to obtain. This work aims to improve the knowledge-learning performance in multi-modal pre-training. Some researchers focus on injecting entity knowledge into language pre-trained models based on masked entity model (MEM) training, which masks entities randomly and lets the model recover. These methods cannot guarantee good performance due to the lack of consideration of which entities are more valuable for learning. Moreover, in multi-modal training data, some entities may be unrelated to visual content. In this work, for the vision-language pre-trained model, we propose a Masked Entity Model pre-training method based on Active learning (ActiveMEM). It is designed to actively mask important and informative entities—those that are both informative and uncertain—for the model to recover, thereby encouraging it to extract more valuable knowledge from the data. The proposed method is evaluated using three pre-training datasets and four downstream datasets, and the experimental results demonstrate the effectiveness of our method.
Förderung
- Shanghai Aerospace Science and Technology Innovation Foundation
- Aviation Science Fund
- Beijing Natural Science Foundation
Zitieren
APA
Wang, Q., Zhang, S., Shao, B., Guo, X., An, J., Fan, C., Jing, Y., Li, J., & Wu, W. (2026). Improving Entity Understanding for Vision-Language Pre-Training via Active Learning. Big Data and Cognitive Computing, 10(6), 198. https://doi.org/10.3390/bdcc10060198
Vancouver
Wang Q, Zhang S, Shao B, Guo X, An J, Fan C, et al. Improving Entity Understanding for Vision-Language Pre-Training via Active Learning. Big Data and Cognitive Computing. 2026;10(6):198. doi:10.3390/bdcc10060198
BibTeX
@article{Wang2026,
author = {Wang, Qunbo and Zhang, Sen and Shao, Boxuan and Guo, Xize and An, Jiayong and Fan, Chao and Jing, Yuanjun and Li, Junxian and Wu, Wenjun},
title = {Improving Entity Understanding for Vision-Language Pre-Training via Active Learning},
journal = {Big Data and Cognitive Computing},
year = {2026},
volume = {10},
number = {6},
pages = {198},
doi = {10.3390/bdcc10060198}
}
EUVIMED