Goto

Collaborating Authors

 physics community


Large Physics Models: Towards a collaborative approach with Large Language Models and Foundation Models

arXiv.org Artificial Intelligence

This paper explores ideas and provides a potential roadmap for the development and evaluation of physics-specific large-scale AI models, which we call Large Physics Models (LPMs). These models, based on foundation models such as Large Language Models (LLMs) - trained on broad data - are tailored to address the demands of physics research. LPMs can function independently or as part of an integrated framework. This framework can incorporate specialized tools, including symbolic reasoning modules for mathematical manipulations, frameworks to analyse specific experimental and simulated data, and mechanisms for synthesizing theories and scientific literature. We begin by examining whether the physics community should actively develop and refine dedicated models, rather than relying solely on commercial LLMs. We then outline how LPMs can be realized through interdisciplinary collaboration among experts in physics, computer science, and philosophy of science. To integrate these models effectively, we identify three key pillars: Development, Evaluation, and Philosophical Reflection. Development focuses on constructing models capable of processing physics texts, mathematical formulations, and diverse physical data. Evaluation assesses accuracy and reliability by testing and benchmarking. Finally, Philosophical Reflection encompasses the analysis of broader implications of LLMs in physics, including their potential to generate new scientific understanding and what novel collaboration dynamics might arise in research. Inspired by the organizational structure of experimental collaborations in particle physics, we propose a similarly interdisciplinary and collaborative approach to building and refining Large Physics Models. This roadmap provides specific objectives, defines pathways to achieve them, and identifies challenges that must be addressed to realise physics-specific large scale AI models.


Physics in the Spotlight

#artificialintelligence

The Institute of Physics (IOP) is convening a new type of forum – four days of discussion, debate and knowledge sharing at the IOP's flagship King's Cross building. Members of IOP special interest groups, including leading thinkers in the physics community in the UK and Ireland, will host discussions, talks, meetings and panels to foster debate and exploration of key topics such as oceans and the climate, optics in quantum technologies and machine learning. IOP member-led special interest groups are at the heart of the physics community and enable you to take part in activities in your particular areas of interest, share and receive expert advice on developments in your sector, grow your network across disciplines, attend topical and inter-disciplinary sessions. A drinks reception will be held at the end of each day, where you can meet speakers and the organising committees.