Abstract: Systems and methods of searching content using one or more machine learning models are disclosed. In response to a user query request, a user query is received at a first user interface. One or more search terms are determined using a query parser based on the user query. One or more semantically similar queries are determined based on the one or more search terms. In accordance with a determination that the user query and the one or more semantically similar queries have not been previously searched, search for information related to the one or more search terms from one or more databases using the one or more machine learning models. An output to the user query is generated based on the information. The output includes at least a portion of the information and one or more citations associated with the information. The output is displayed at the first user interface.