（五十六）：Integrating Multimodal Information in Large Pretrained Transformers
- Abstract
- 1. Introduction
- 2. Related Work
-
- 2.1多模态语言分析
- 2.2预训练的语言表征
- 3 BERT and XLNet
-
- 3.1 Transformer
- 3.2 Transformer-XL
- 3.3 BERT
- 3.4 XLNet
- 4 Multimodal Adaptation Gate (MAG)
-
- 4.1 MAG-BERT
- 4.2 MAG-XLNet
- 5 Experiments
-