Described embodiments provide systems and methods for managing session accessed by a client device. The systems and methods can include one or more processors configured to receive data in a plurality of modalities corresponding to a plurality of features of a session for an entity accessed by a client device. The one or more processors can determine based on the data of the session and a distance model trained with historical data of the entity, a distance between a representation of the data of the session and a predetermined representation for the entity established based on the historical data of the entity. The one or more processors can compare the distance with a threshold established for the entity. The one or more processors can generate, based on the comparison between the distance with the threshold, an action to manage access by the client device to the session for the entity.