There are multiple ways to extract values and entities from user utterance. The following video shows how to provide a few slot filling examples in PromptDialog so that value extraction and/or entity recognition models can be learned in RASA.