We study classical background electric fields and the Schwinger effect in de Sitter space. We show that a constant electric field in de Sitter requires the photon to have a tachyonic mass proportional to the Hubble scale. This has physical implications for the induced Schwinger current that affect its IR behaviour. To study this, we recompute the Schwinger current in de Sitter space for charged fermions and minimally coupled scalars, imposing a physically consistent renormalization condition. We find a $\textit{finite and positive}$ Schwinger current even in the massless limit. This is in contrast to previous calculations in the literature, which found a negative IR divergence. We also obtain the first result for the Schwinger current of a non-minimally coupled scalar, including the conformally coupled case, which we find exhibits behaviour very similar to that of the fermion current. Our results may have physical implications for both magnetogenesis and inflationary dark matter production.

Spain

Mar Bastero-Gil

Portugal

Paulo B. Ferraz

António Torres Manso

Slovenia

Italy

Lorenzo Ubaldi

Roberto Vega-Morales

Schwinger current in de Sitter space * This work has been partially supported by Junta de Andalucía Projects P21-00199 and A-FQM-472-UGR20 (FEDER funds); SRA (10.13039/501100011033) and ERDF (PID2022-139466NB-C21) (R.V.M.); PID2022-140831NB-I00, funded by MICIU/AEI/10.13039/501100011033 and FEDER, UE (M.B.G., A.T.M.); by FCT CERN (10.54499/2024.00252.CERN) (M.B.G., A.T.M., P.F.); FCT Grant No. SFRH/BD/151475/2021 and 10.54499/SFRH/BD/151475/2021 (P.F.). LU is supported by the Slovenian Research Agency under research core funding No. P1-0035 and by research grants J1-60026 and J1-4389. This article is based upon work from COST Action COSMIC WISPers CA21106, supported by COST (European Cooperation in Science and Technology). R.V.M. thanks the Mainz Institute for Theoretical Physics (MITP) of the Cluster of Excellence PRISMA$^{+}$ (Project ID 390831469) for its hospitality and partial support during the completion of this work

Particle physics experiments rely on the (generalised) likelihood ratio test (LRT) for searches and measurements, which consist of composite hypothesis tests. However, this test is not guaranteed to be optimal, as the Neyman–Pearson lemma pertains only to simple hypothesis tests. Any choice of test statistic thus implicitly determines how statistical power varies across the parameter space. An improvement in the core statistical testing methodology for general settings with composite tests would have widespread ramifications across experiments. We discuss an alternate test statistic that provides the data analyzer an ability to focus the power of the test on physics-motivated regions of the parameter space. We demonstrate the improvement from this technique compared to the LRT on a Higgs $$\rightarrow \tau \tau $$ dataset simulated by the ATLAS experiment and a dark matter dataset inspired by the LZ experiment. We also employ machine learning to efficiently calibrate critical values for a family of tests, which are then inverted to obtain statistically valid confidence intervals.

United States

James Carzon

Aishik Ghosh

Brazil

Rafael Izbicki

Ann Lee

Luca Masserano

Daniel Whiteson

On focusing statistical power for searches and measurements in particle physics

Hard scattering events in high-energy collisions produce highly virtual partons that subsequently fragment into collimated hadronic cascades. When such partonic showers evolve in a QCD medium, as in deep-inelastic scattering or heavy-ion collisions, the resulting multi-particle distributions encode information about the surrounding matter. Decades of theoretical developments have led to a consistent and order-by-order improvable perturbative description of the shower. This description needs, however, the non-perturbative input that encodes the structure of the hadronic matter. The determination of such input remains challenging within conventional computational approaches, thereby limiting the applicability of the approach. In this work, we develop a framework that employs quantum simulation techniques to compute multi-particle processes in such environments by mapping partonic cross-sections to quantum circuits. As benchmarks, we analyze dipole formation and the QCD antenna radiation pattern at leading order in the strong coupling constant, comparing the results with analytic estimates in simplified limits. The quantum circuit formulation here introduced naturally extends to higher perturbative orders and enables amplitude-level computations in complex matter backgrounds. This provides a systematic foundation for applying quantum information science methods to study multi-particle dynamics in QCD media.

Switzerland

João Barata

Meijian Li

China

Wenyang Qian

Carlos Salgado

João Silva

Quantum simulating multi-particle processes in high energy nuclear physics: dijet production and color (de)coherence

Unsupervised learning has been widely applied to various tasks in particle physics. However, existing models lack precise control over their learned representations, limiting physical interpretability and hindering their use for accurate measurements. We propose the histogram autoencoder (HistoAE), an unsupervised representation learning network featuring a custom histogram-based loss that enforces a physically structured latent space. Applied to silicon microstrip detectors, HistoAE learns an interpretable two-dimensional latent space corresponding to the particle’s charge and impact position. After simple postprocessing, it achieves a charge resolution of <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mn>0.25</mn><mi>e</mi></math> and a position resolution of <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mn>3</mn><mtext> </mtext><mtext> </mtext><mi mathvariant="normal">μ</mi><mrow><mi mathvariant="normal">m</mi></mrow></math> on beam-test data, comparable to the conventional approach. These results demonstrate that unsupervised deep learning models can enable physically meaningful and quantitatively precise measurements. Moreover, the generative capacity of HistoAE enables straightforward extensions to fast detector simulations.

Xing-Jian Lv

De-Xing Miao

Zi-Jun Xu

Jian-Chun Wang

Interpretable unsupervised representation learning for high-precision measurements in particle physics

We propose a nonperturbative completion of two-point correlators in <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mi>T</mi><mover accent="true"><mi>T</mi><mo stretchy="false">¯</mo></mover></math>-deformed conformal field theories (CFTs) and analyze their behavior at distance scales shorter than the fundamental length scale set by the <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mi>T</mi><mover accent="true"><mi>T</mi><mo stretchy="false">¯</mo></mover></math> deformation. Building on the interpretation of the <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mi>T</mi><mover accent="true"><mi>T</mi><mo stretchy="false">¯</mo></mover></math> deformation as a coupling to two-dimensional quantum gravity with a unique built-in length scale, we advance the study of <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mi>T</mi><mover accent="true"><mi>T</mi><mo stretchy="false">¯</mo></mover></math>-deformed CFTs as a toy model for Planckian physics. As we probe shorter distances, trans-Planckian oscillations are followed by a super-Planckian regime in which correlations are typically suppressed by geometric randomness, in contrast to the power-law growth characteristic of CFTs. Moreover, their dependence on distance becomes exponentially weaker, suggesting that the underlying geometric structure has been largely erased—a behavior broadly consistent with expectations for quantum spacetime in this regime.

Japan

Shinji Hirano

Vinayak Raj

Nonperturbative effects in <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mi>T</mi><mover accent="true"><mi>T</mi><mo stretchy="false">¯</mo></mover></math>-deformed conformal field theories: A toy model for Planckian physics

We present a machine-learning-based anomaly detection strategy designed to identify anomalous physics in events containing resonant Standard Model physics and demonstrate this method on the final state of a Higgs boson decaying to two photons. The demonstration targets high-dimensional deviations in the region of phase space containing the Higgs mass peak in a fully signal-agnostic manner. A latent-space embedding, learned from event kinematics, enables the use of a large set of potentially sensitive features. Backgrounds are estimated using a hybrid approach that combines machine-learning-based generative modeling with traditional simulation, and a discriminator is trained in the latent space to distinguish data from background estimates. After applying a selection on the classifier output, the invariant mass distribution of the diphoton system is examined for localized excesses above the simulated Higgs peak. We benchmark the sensitivity of this strategy using simplified simulated proton-proton collisions corresponding to data recorded during Run 2 of the LHC and show that the method can provide significant improvements in sensitivity, even for small signal injections that could remain undetected in an inclusive analysis. These results demonstrate that the proposed strategy is a promising and viable approach for future searches and should be applied to recorded collider data.

Chi Lung Cheng

Sarah Demers

Sascha Diefenbacher

Runze Li

Benjamin Nachman

Dennis Noll

Weakly supervised anomaly detection in events with a Higgs boson and exotic physics

Hao-Fu Zhu

Guo-Zhu Liu

Xufen Wu

Corrigendum to “Rapid cooling of the Cassiopeia A neutron star due to superfluid quantum criticality” [Physics Letters B 879 (2026) 140591]

Given the non-negligible interplay between parton distribution functions (PDFs) at large $\textit{x}$ and potential New Physics (NP) effects in the high-energy tails of hadron collider observables, a central question is which PDFs can be reliably employed in beyond-the-Standard-Model (BSM) analyses. In this work, we examine the fine balance between using PDF sets with small uncertainties in the large-$\textit{x}$ region — crucial for maximising BSM sensitivity — and adopting conservative PDF fits that exclude high-energy data potentially contaminated by unaccounted NP contributions. We systematically assess a range of conservative PDF fitting strategies designed to mitigate such biases and provide a recommendation for the class of PDFs best suited for robust BSM searches. In addition, we investigate the alternative approach of performing simultaneous fits of Standard Model Effective Field Theory (SMEFT) Wilson coefficients and PDFs, thereby consistently accounting for their mutual correlations. Starting from a toy model to illustrate the underlying mechanisms, we then analyse two realistic NP scenarios: one modifying high-mass Drell-Yan production and another affecting the high-invariant-mass tail of top-quark pair production. Both cases are representative of measurements that will be probed with high precision during the High-Luminosity phase of the LHC.

United Kingdom

Ella Cole

Mark Costantini

Netherlands

Elie Hammou

Luca Mantani

Francesco Merlotti

Manuel Morales-Alvarado

Maria Ubiali

Tailored PDFs for new physics searches

Density functional theory is a practical approach for solving quantum many-body problems with available computational resources. The complexity of the nuclear force makes constructing an accurate nuclear energy density functional much more challenging. The feasibility of constructing a nuclear covariant energy density functional with deep neural networks is demonstrated. This physics-guaranteed neural network approach achieves high accuracy in predicting nuclear energy density and exhibits significantly better extrapolation abilities than traditional machine learning methods for binding energies. When combined with the existing covariant density functional, the neural network approach improves the binding energy accuracy from 644 keV to 86 keV in the known region and also effectively captures the microscopic shell effect. Furthermore, its extrapolation performance is also significantly enhanced, achieving an accuracy of approximately 5 MeV even when extrapolating up to 30 steps. This work paves the way for the construction of accurate nuclear energy density functionals through machine learning.

W.F. Li(李伟峰)

Z.M. Niu(牛中明)

H.Z. Liang(梁豪兆)

Y.F. Niu(牛一斐)

B.H. Sun(孙保华)

Construction of nuclear covariant energy density functional from a physics-guaranteed neural network approach

Recent studies have shown that quantitative concepts from quantum information theory can play a role in analyzing collider physics, including elucidating new physics. In this paper, we study various quantum information measures including magic, trace distance, and fidelity distance, in generic new physics scenarios modeled by the standard model effective field theory. We argue that such measures can indeed show up differences with respect to the pure standard model, and we compare our results with similar findings for the concurrence discussed previously in the literature. We examine the relative sensitivity of different measures to new physics in two-dimensional bins of the top pair invariant mass and scattering angle, finding that the concurrence, magic, and trace distance each emerge as the best measure in at least some regions of the phase space. This highlights the importance of exploring multiple quantum information measures in the hunt for beyond the standard model physics.