DepthViT, a ~300k-parameter masked-autoencoder ensemble with depth-wise attention and per-run refreshed Z-statistics, sustains >98.8% precision on synthetic HCAL occupancy anomalies across CMS 2018/2022 runs.
Data Quality Monitoring for the Hadron Calorimeters Using Transfer Learning for Anomaly Detection,
1 Pith paper cite this work, alongside 2 external citations. Polarity classification is still indexing.
1
Pith paper citing it
2
external citations · OpenAlex
fields
hep-ex 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Continual Learning via Ensemble-Based Depth-Wise Masked Autoencoders for Data Quality Monitoring in High-Energy Physics
DepthViT, a ~300k-parameter masked-autoencoder ensemble with depth-wise attention and per-run refreshed Z-statistics, sustains >98.8% precision on synthetic HCAL occupancy anomalies across CMS 2018/2022 runs.