Large Physics Models: Towards a collaborative approach with Large Language Models and Foundation Models
Authors:
Kristian G. Barman,
Sascha Caron,
Emily Sullivan,
Henk W. de Regt,
Roberto Ruiz de Austri,
Mieke Boon,
Michael Färber,
Stefan Fröse,
Faegheh Hasibi,
Andreas Ipp,
Rukshak Kapoor,
Gregor Kasieczka,
Daniel Kostić,
Michael Krämer,
Tobias Golling,
Luis G. Lopez,
Jesus Marco,
Sydney Otten,
Pawel Pawlowski,
Pietro Vischia,
Erik Weber,
Christoph Weniger
Abstract:
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…
▽ More
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.
△ Less
Submitted 9 January, 2025;
originally announced January 2025.
Quantum effects in topological and Schwarzschild de Sitter brane: Aspects of torsion on ${\mathbf{(D{\bar D})_4}}$-brane universe
Authors:
Richa Kapoor,
Supriya Kar,
Deobrat Singh
Abstract:
We investigate an effective torsion curvature in a second order formalism underlying a two form world-volume dynamics in a $D_5$-brane. In particular, we consider the two form in presence of a background (open string) metric in a $U(1)$ gauge theory. Interestingly the formalism may be viewed via a non-coincident pair of $(D{\bar D})_5$-brane with a global NS two form on an anti brane and a local t…
▽ More
We investigate an effective torsion curvature in a second order formalism underlying a two form world-volume dynamics in a $D_5$-brane. In particular, we consider the two form in presence of a background (open string) metric in a $U(1)$ gauge theory. Interestingly the formalism may be viewed via a non-coincident pair of $(D{\bar D})_5$-brane with a global NS two form on an anti brane and a local two form on a brane. The energy-momentum tensor is computed in the six dimensional CFT. It is shown to source a metric fluctuation on a vacuum created pair of $(D{\bar D})_4$-brane at a cosmological horizon by the two form quanta in the gauge theory. The emergent gravity scenario is shown to describe a low energy (perturbative) string vacuum in $6D$ with a (non-perturbative) quantum correction by a lower ($p<5$) dimensional $D_p$ brane or an anti brane in the formalism. A closed string exchange between a pair of $(D{\bar D})_4$-brane, underlying a closed/open string duality, is argued to describe the Einstein vacuum in a low energy limit. We obtain topological de Sitter and Schwarzschild brane universe in six dimensions. The brane/anti-brane geometries are analyzed to explore some of their characteristic and thermal behaviours in presence of the quantum effects. They reveal an underlying nine dimensional type IIA and IIB superstring theories on $S^1$.
△ Less
Submitted 18 October, 2014; v1 submitted 29 July, 2014;
originally announced July 2014.
Quintessence and effective RN de Sitter brane geometries
Authors:
K. Priyabrat Pandey,
Abhishek K. Singh,
Sunita Singh,
Richa Kapoor,
Supriya Kar
Abstract:
We revisit an effective space-time torsion curvature in a second order formalism, underlying the non-linear $U(1)$ gauge dynamics, of a two form on a $D_4$-brane in type IIA superstring theory. The formalism incorporates the significance of a global NS two form into the theory via its perturbative coupling to a dynamical two form. In particular, we explore the non-linear gauge dynamics on a $D_4$-…
▽ More
We revisit an effective space-time torsion curvature in a second order formalism, underlying the non-linear $U(1)$ gauge dynamics, of a two form on a $D_4$-brane in type IIA superstring theory. The formalism incorporates the significance of a global NS two form into the theory via its perturbative coupling to a dynamical two form. In particular, we explore the non-linear gauge dynamics on a $D_4$-brane in presence of a non-trivial background metric. The fact that the global modes of a NS two form in an open string theory sources the background metric on a $D_4$-brane may hint at the existence of an anti $D_4$-brane in the formalism. An effective de Sitter universe is shown to emerge on a vacuum created pair of $(D{\bar D})_3$-brane by a local two form at the past horizon with a Big Bang. We obtain a number of 4D de Sitter quantum black holes, including a Reissner-Nordstrom vacuum, with and without a propagating torsion. The quantum black holes are shown to be free from curvature singularity at $r\rightarrow 0$. In a low energy limit, the nonperturbative correction sourced by a torison may seen to be insignificant. The quantum black hole undergoes an expansion in the limit and presumably identifies with the Einstein vacuum. Interestingly our analysis reveals a plausible quintessence (axion) on an anti $D_3$-brane which may source the dark energy in a $D_3$-brane universe. Arguably a brane universe moves away from its anti-brane due to the conjectured repulsive gravity underlying the quintessence. It leads to a growth in extra fifth dimension between a brane and an anti-brane which may provide a clue behind an accelerating universe observed in cosmology.
△ Less
Submitted 15 May, 2014;
originally announced May 2014.
