Update README.md

README.md

@@ -22,6 +22,7 @@ Kontext dev LoRA finetuned for extract—clothes images, trained on a dataset of
 
 ## Trigger words
 
-Replace "clothes" with the corresponding type of clothing, such as: shirt, trouser, hat, coat, etc.
+Replace "**clothes**" with the corresponding type of clothing, It is best to extract single pieces of clothing.such as: shirt, trouser, hat, coat, etc.
 
-extract only the `clothes` over a white background, product photography style
+
+extract only the **clothes** over a white background, product photography style