Update readme.md

Author: Peng99999

readme.md

@@ -1,6 +1,7 @@
 We list the OpenSource Datasets in six types:
 
-**1.Texts**
+**1.Texts**:
+1. [GLUE (General Language Understanding Evaluation benchmark)](https://github.com/nyu-mll/GLUE-baselines) General Language Understanding Evaluation (GLUE) benchmark is a collection of nine natural language understanding tasks, including single-sentence tasks CoLA and SST-2, similarity and paraphrasing tasks MRPC, STS-B and QQP, and natural language inference tasks MNLI, QNLI, RTE and WNLI.
 
 **2.Images**