diff --git a/README.md b/README.md
index 3ff4dcc..42c1525 100644
--- a/README.md
+++ b/README.md
@@ -4,7 +4,7 @@ This is the implementation of the paper [Meta Self-training for Few-shot Neural
 
 
 ## Overview
-![](./figs/metast.pdf)
+![](./figs/metast.png)
 
 In this work we present MetaST, a self-training framework with auto data selection technique to prevent error probagation.
 You can find more details of this work in our [paper](https://dl.acm.org/doi/pdf/10.1145/3447548.3467235).