Transformer-Based Analysis for Next-Generation Compact Cherenkov Telescopes

The past decade has witnessed a rapid expansion of machine learning applications in very-high-energy (VHE) gamma-ray astronomy. While many efforts to apply deep learning to IACT data analysis have focused on Convolutional Neural Networks (CNNs), these approaches remain constrained by irregular camera geometries and the limited ability of Monte Carlo simulations to accurately model existing telescopes. In this work, we employ a fully virtual simulated telescope optimization framework to investigate how advanced machine learning techniques can fundamentally reshape instrumental design. Our analysis achieves an order-of-magnitude improvement in performance compared to a typical standard analysis pipeline. Within this framework, a single 5-meter-diameter IACT can reach an energy threshold one order of magnitude lower than standard methods. We furthermore find that by lowering the detection efficiency of the telescope camera, the deep-learning analysis suffers less, and we are able to achieve performance comparable, within a factor of a few, to that of the LST-1 telescope, a 23m-diameter IACT. These findings point toward a viable pathway for the development of cost-effective, potentially autonomous IACTs, capable of substantially increasing the temporal coverage and duty cycle of future gamma-ray observatories.