LHC Truth-Jet and Reco-Jet Datasets for Parity Violation Studies

Simulated datasets from the Large Hadron Collider used to hunt for parity-violating signals beyond the Standard Model. The data, created by C. G. Lester at the University of Cambridge, contains reconstructed four-momenta for the five hardest jets per event, stored in HDF5 files with shape (n, 20). Truth-jet files include additional truth-level flavour and helicity information encoded using PDG IDs.

Use Cases

• Training anomaly detection models based on simulated jet four-momenta to identify parity-violating signals.
• Comparing Standard Model simulations against parity-violating mSME model simulations based on jet data.
• Studying detector rotation effects on signal reconstruction using the rotated dataset subset.
• Developing classifiers for jet flavour and helicity based on the truth-level labels provided.

Strengths

• Includes independent train, test, and private_test splits, with the private_test set reserved for final model evaluation.
• Truth-jet files provide two additional structured data keys for flavour and helicity information.
• Contains rotated dataset variants to study systematic effects from detector orientation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

University of Cambridge

Collection Method

Processed results from particle physics simulations at the Large Hadron Collider.

Time Range

null

Freshness

null

Geography

null