(四十九):UNIMO: Towards Unified-Modal Understanding and Generation via Cross-Modal Contrastive Learning
- Abstract
- 1 Introduction
- 2 UNIMO
-
- 2.1 Cross-Modal Contrastive Learning
-
- 文本重写
- 图像/文本检索
- 2.2 Visual Learning
- 2.3 Language Learning
-
- 双向预测
- Seq2Seq生成
- 3 Experimental Settings
-
- 3.1 Pre-training Dataset
- 3.2 Implementation Detail