Reasoning AIccelerators for High Energy Physics
Nathaniel Craig
University of California, Santa Barbara (UCSB)
Physics
Theoretical breakthroughs in high-energy physics are often driven by acquiring qualitatively new data and making optimal use of the data at hand. While this has traditionally been accomplished with particle accelerators, reasoning models hold great promise as AI accelerators: generators of 'theoretical data’ and interpreters of theoretical and experimental data alike. I’ll survey some of the open questions facing high energy physics and present six aspirational targets for reasoning models, spanning sixty orders of magnitude from the Hubble scale to the Planck scale. I'll then demonstrate the current power of reasoning models with the FERMIAcc, a particle theory agent for interpreting collider data.
Back to Long Programs