Update README.md

README.md

@@ -41,6 +41,8 @@ Due to compute constraint, we selected only single model to compare against - no
 
 Future benchmark iterations might include different arches, more models and more characters.
 
+P.S. While benchmark tests characters purely with their own trigger with no helper tags, it is advised to utilize series/game for better adherence. Characters that might appear not working initially could start working with appearance tags.
+
 ## Bias and Limitations
 
 General data biases from Danbooru might apply.