Abstract: A machine learning (ML) based web application firewall (WAF) is described. Transformation(s) are applied to raw data including normalizing and generating a signature over the normalized data. The signature and the normalized data are vectorized to create a first and second vector of integers that are input into an ML model that includes a first stage that operates on the first vector of integers to identify candidate signature tokens that are commonly associated with different classes of attack, and a second stage that operates on the candidate signature tokens and the second vector of integers and conditions attention on the second vector of integers on the candidate signature tokens. The ML model outputs a score that indicates a probability of the raw data being of a type that is malicious. A traffic processing rule is enforced that instructs a WAF to block traffic when the score is above a threshold.

Abstract: Systems and methods for linking real world activity with a location-based parallel reality experience are provided. In particular, a game server hosting a parallel reality experience receives position data from a client device of a user and determines whether the client device is at a real world landmark associated with a geographic area. Responsive to determining that the client device is at the real world landmark, the game server cause the client device to present content incentivizing the user to travel to a real world location in the geographic area. The game server determines whether the client device is within proximity of the real world location based on sensor data. Responsive to determining that the client device is within proximity of the real world location, the game server causes the client device to display the virtual element at a virtual world location corresponding to the real world location.