Takahiro Kawahara | Searcharxiv

arXiv2026

Jet flavor tagging with Particle Transformer for Higgs factories

We study the performance of the Particle Transformer (ParT) for jet flavor tagging using ILD full simulation events (1M jets) as well as fast simulation samples (10M and 1M jets). We perform 3-category ($b/c/d$), 6-category ($b/c/d/u/s/g$), and 11-category trainings (including quark--antiquark separation), incorporating multivariate hadron particle identification information from $dE/dx$ and time-of-flight. For $b$/$c$ tagging, we observe a factor of 5--10 improvement over previous BDT-based taggers, and we obtain reasonable performance for strange tagging and quark/antiquark separation. The 10M-jet fast simulation study indicates that further gains are possible with higher training statistics.

physics.data-an↗

arXiv2026

Search for Sub-GeV Axion-Like Particles at EBES Pilot Run Using 4 GeV Positron Beam at KEK LINAC

We report the results of a search for sub-GeV axion-like particles (ALPs) using pilot run data from the Electron Beam-dump Experiment at KEK LINAC Switching Yard 3 (EBES). The data were collected in December 2023 with a 4 GeV positron beam and correspond to $1.3\times10^{14}$ positrons on target. In the pilot run setup, a tungsten beam dump and a single PbO calorimeter were used. We consider ALP production via the Primakoff process induced by bremsstrahlung photons in the beam dump, followed by the decay $a\toγγ$. The background was estimated with a data-driven method, and a signal region was defined such that the expected background yield is below 0.1 events. No events were observed after unblinding. Upper limits at the 90% confidence level were derived in the ALP mass-coupling plane, extending the experimental coverage into a region of parameter space not explored by previous searches.

hep-ex↗