In this study, we introduce the More-Interaction Particle Transformer (MIParT), a novel deep learning neural network designed for jet tagging. This framework incorporates our own design, the More-Interaction Attention (MIA) mechanism, which increases the dimensionality of particle interaction embeddings. We tested MIParT using the top tagging and quark-gluon datasets. Our results show that MIParT not only matches the accuracy and AUC of LorentzNet and a series of Lorentz-equivariant methods, but also significantly outperforms the ParT model in background rejection. Specifically, it improves background rejection by approximately 25% at a 30% signal efficiency on the top tagging dataset and by 3% on the quark-gluon dataset. Additionally, MIParT requires only 30% of the parameters and 53% of the computational complexity needed by ParT, proving that high performance can be achieved with reduced model complexity. For very large datasets, we double the dimension of particle embeddings, referring to this variant as MIParT-Large (MIParT-L). We find that MIParT-L can further capitalize on the knowledge from large datasets. From a model pre-trained on the 100M JetClass dataset, the background rejection performance of the fine-tuned MIParT-L improved by 39% on the top tagging dataset and by 6% on the quark-gluon dataset, surpassing that of the fine-tuned ParT. Specifically, the background rejection of fine-tuned MIParT-L improved by an additional 2% compared to the fine-tuned ParT. The results suggest that MIParT has the potential to advance efficiency benchmarks for jet tagging and event identification in particle physics. The code is available at the following GitHub repository: https://github.com/USST-HEP/MIParT
Abstract Background Conservation of resources theory (COR) establishes a link between resource loss and the stress response. The aim of this study was to assess the contribution of resource loss in the form of home damage and the choice of active or passive coping strategies to PTSD symptoms in survivors of the 2020 Petrinja (Croatia) earthquake. Methods A total of 374 adults (29.9% men) aged 18–64 years living in the counties surrounding the epicenter of the Petrinja (Croatia) earthquake participated in an online cross-sectional survey. The questionnaire included the PTSD Checklist for DSM-5 (PCL-5), the Coping Inventory, and the binary item assessing whether or not the participants' home was damaged. Results Hierarchical regression analysis showed that home damage was a significant predictor of PTSD symptoms. Participants whose homes were damaged by the earthquake were significantly more likely to use passive coping strategies, namely avoidance and emotional venting, and one active coping strategy, action, than those whose homes were spared. Finally, more frequent use of passive coping was associated with a higher risk of PTSD symptoms. Conclusions The study corroborates the COR theory link between resource loss and the stress response, as well as the general consensus that passive coping is a less adaptive strategy than active coping. In addition to passive coping, individuals who lacked resources may have been inclined to take some active steps because they either needed to repair or relocate their homes and because most buildings were only moderately to minimally damaged in the Petrinja earthquake.
In this talk, we present a way to improve the accuracy of theoretical predictions for Higgs boson production cross sections at the LHC using the measurements of lepton angular distributions. In this regards, we exploit the sensitivity of the lepton angular coefficient associated with the longitudinal Z-boson polarization to the parton density function (PDF) for gluons resolved from the incoming protons, in order to constrain the Higgs boson cross section from gluon fusion processes. We find that high-statistics determinations of the longitudinally polarized angular coefficient at the LHC Run 3 and high-luminosity HL-LHC improve the PDF systematics of the Higgs boson cross section predictions by 50% over a broad range of Higgs boson rapidities. This study has been conducted using the open-source fitting framework xFitter. This talk refers to the following paper: arXiv:2012.10298[hep-ph]
In this paper, as the second part of the third step of our study on developing data analysis procedures for using 3-dimensional information offered by directional direct Dark Matter detection experiments in the future, we investigate the angular distributions of the recoil direction (flux) and the recoil energy of the Monte Carlo simulated WIMP-scattered target nuclei observed in different celestial coordinate systems. The "anisotropy" and the "directionality" ("annual" modulation) of the angular recoil-direction/energy distributions will be demonstrated. We will also discuss their dependences on the target nucleus and on the mass of incident halo WIMPs. For readers' reference, all simulation plots presented in this paper (and more) can be found "in animation" on our online (interactive) demonstration webpage (http://www.tir.tw/phys/hep/dm/amidas-2d/).
Azithromycin (AZI) has been listed as an emerging contaminant by the US EPA since 2009 because it is frequently detected in wastewater, surface water, and drinking water. In this paper, the chlorination of AZI in drinking water was simulated and studied. The results indicated that new compounds were generated in the chlorination of AZI. The byproducts were identified by liquid chromatography-quadrupole-time-of-flight mass spectrometry (LC-Q-TOF), and four of the byproducts were detected in real water samples. The kinetic studies demonstrated that the reaction rates of AZI chlorination were dependent on the initial concentration of free chlorine and the pH value. The potential toxicities of the byproducts were assessed by quantitative structure-activity relationship software and investigated by the viability of Chinese hamster lung (CHL), Jurkat T and Hep G2 cells.
We have presented the results of our next-to-next-to-leading order (NNLO) QCD analysis of nuclear parton distribution functions (nuclear PDFs) [Phys. Rev. D 93 (2016) 014026, arXiv:1601.00939 [hep-ph]] using all available neutral current charged-lepton ($\ell ^\pm$ + nucleus) deeply inelastic scattering (DIS) data as well as Drell-Yan (DY) cross-section ratios $\sigma_{\rm DY}^{A}/\sigma_{\rm DY}^{A^\prime}$ for a variety of nuclear targets. We have studied in detail the parametrizations and the atomic mass (A) dependence of the nuclear PDFs at NNLO at the input scale, $Q_0^2 = 2 \, {\rm GeV^2}$. Our {\tt KA15} nuclear PDFs provides a complete set of nuclear PDFs, $f_i^{(A, Z)}(x, Q^2)$, with a full functional dependence on $x$, A, $Q^2$. The uncertainties of the obtained nuclear modification factors for each parton flavour as well as the corresponding observables are estimated using the well-known Hessian method. The nuclear heavy quark contributions are also included into the analysis in the framework of zero-mass variable flavour number scheme (ZM-VFNS). We compare the results of our parametrization with all available nuclear DIS data and the results of other nuclear PDFs groups. We have found that our nuclear PDFs to be in reasonably good agreement with results in the literature. The estimates of errors provided by our global analysis ({\tt KA15}) are rather smaller than those of other groups. We have briefly reviewed different aspects of recent LHC heavy-ion collisions data including the first experimental data from the LHC proton+lead ($p-pb$) and lead+lead ($pb-pb$) run which can be used in the global fits of nuclear PDFs.
In this talk, I present a novel and minimal alternative to thermal leptogenesis, which builds upon the assumption that the electroweak gauge bosons are coupled to an axion-like scalar field, as it is, for instance, the case in certain string compactifications. The motion of this axion-like field after the end of inflation generates an effective chemical potential for leptons and antileptons, which, in the presence of lepton number-violating scatterings mediated by heavy Majorana neutrinos, provides an opportunity for baryogenesis via leptogenesis. In contrast to thermal leptogenesis, the final baryon asymmetry turns out to be insensitive to the masses and CP-violating phases in the heavy neutrino sector. Moreover, the proposed scenario requires a reheating temperature of at least O(10^12) GeV and it is, in particular, consistent with heavy neutrino masses close the scale of grand unification. This talk was given in February 2015 at HPNP 2015 at Toyama University and is based on recent work (arXiv:1412.2043 [hep-ph]) in collaboration with A. Kusenko and T. T. Yanagida.