A detector's full record can tell more than the single number it is usually boiled down to
A new preprint finds that using the whole simulated record of a particle shower sharpens the energy reading by a factor of 1.7 to 1.9.
Picture made for ABUREAU.
A single-author preprint, Calorimetry as inference, submitted on 30 October 2026, asks how much a particle detector's record really holds. A calorimeter infers a particle's energy from the trace left by a shower. The usual method boils that trace down to one summed number.
The author tests this on simulated hadronic showers in copper, between 20 and 100 GeV. Using the full record, the paper reports an energy resolution 1.7 to 1.9 times better than the standard single-number construction on the same events. Dual-readout channels give the largest share of the gain, and timing adds about as much as shower shape.
The data is simulated, not measured, and the paper says some of the gain depends on the simulation physics. The same records, read more fully, give a better answer. For anyone who holds data, the point is that value can sit in what is already stored, not only in new collection. The page links no dataset or code of its own.