提交 e4864d9f 编写于 作者: T tianxin

update README

上级 6165f82c
...@@ -36,6 +36,8 @@ We construct several tasks to capture different aspects of information in the tr ...@@ -36,6 +36,8 @@ We construct several tasks to capture different aspects of information in the tr
- **Structure-aware Tasks**: to capture the syntactic information - **Structure-aware Tasks**: to capture the syntactic information
- **Semantic-aware Tasks**: in charge of semantic signals - **Semantic-aware Tasks**: in charge of semantic signals
At the same time, ERINE 2.0 feeds task embedding to model the characteristic of different tasks. We represent different tasks with an ID ranging from 0 to N. Each task ID is assigned to one unique task embedding.
![ernie2.0_model](.metas/ernie2.0_model.png) ![ernie2.0_model](.metas/ernie2.0_model.png)
......
...@@ -36,6 +36,8 @@ ...@@ -36,6 +36,8 @@
- **Structure-aware Tasks**: 语法 (syntactic) 级别信息的学习 - **Structure-aware Tasks**: 语法 (syntactic) 级别信息的学习
- **Semantic-aware Tasks**: 语义 (semantic) 级别信息的学习 - **Semantic-aware Tasks**: 语义 (semantic) 级别信息的学习
同时,针对不同的 pre-training 任务,ERNIE 2.0 引入了 Task Embedding 来精细化地建模不同类型的任务。不同的任务用从 0 到 N 的 ID 表示,每个 ID 代表了不同的预训练任务。
![ernie2.0_model](.metas/ernie2.0_model.png) ![ernie2.0_model](.metas/ernie2.0_model.png)
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册