Update README.md

README.md

@@ -11,6 +11,7 @@ size_categories:
 
 # CP-Bench: A dataset for evaluating LLM-driven constraint modelling
 
-TODO
+TODO: organize the structure
 
-Feel free to contact us if you find any errors in models/solutions/descriptions, the dataset was mostly manually constructed so mistakes are more than possible to exist, thank you!
+* For evaluating: you can use any modelling language you want, as long as the generated final solution is in the required structure.
+* Feel free to contact us if you find any errors in models/solutions/descriptions, the dataset was mostly manually constructed, so mistakes are more than possible to exist, thank you!