Third-year PhD student Qiyuan Xu, from the QUASAR Group and the LIV.INNO Centre for Doctoral Training, was awarded third place in the poster competition at “AI Discovery Day: Powering Research & Innovation with NVIDIA and Dell Technologies,” held on 22 October 2025 at The Spine Building, Liverpool.
The event brought together researchers and industry specialists to explore practical, high-impact uses of AI and high-performance computing in areas such as simulation, robotics and generative AI. Participants were invited to present their work in a way that is directly relevant to industry, with posters judged by industry professionals and academic leads, and prizes awarded to the top three entries.
Qiyuan’s poster presented ongoing work on multimode-fiber-based, machine-learning reconstruction of transverse beam distribution for beam diagnostics. The aim is to recover high-quality transverse beam information from fiber-delivered optical signals and move toward radiation-tolerant diagnostics that can operate in harsh accelerator environments. This work is being developed in collaboration between the University of Liverpool and CERN. On site, the poster was set up and presented by QUASAR Group colleagues Debdeep Ghosal and Alex Hill.
Qiyuan’s poster on multimode-fiber-based, machine-learning reconstruction of transverse beam distribution for beam diagnostics.
Qiyuan commented: “This was a very encouraging result for me. Industry experts were not only interested in the approach we’re developing for accelerator diagnostics but also gave useful feedback on how similar techniques could be transferred to wider industrial monitoring and inspection. That perspective is very valuable for shaping the future direction of the work.”