Hi Pupil Labs community! With some delay, I am still very excited to share the first paper in which my team and I used Neon. In this work, we test the capabilities of vision-language foundation models to simulate human visual attention in free-viewing and captioning. Neon helped us collect reference human gaze data to compare with the mouse-contingent crowd-sourced dataset. https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=11199898