Update README.md

README.md

@@ -28,7 +28,7 @@ It achieves the following results on the evaluation set:
 
 Trained on a summarization task with _potentially_ multiple doc-summary pairs stacked on top of each other.
 
-You can separate it's predictions by using it's special token `[NEXT_CONCEPT]` to split the output into "separate topics".
+You can separate its predictions by using it's special token `[NEXT_CONCEPT]` to split the output into "separate topics".
 
 ## Intended use & limitations