Pith. sign in

REVIEW 35 cited by

ParticleNet: Jet Tagging via Particle Clouds

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1902.08570 v3 pith:34TVFDD5 submitted 2019-02-22 hep-ph cs.CVhep-ex

ParticleNet: Jet Tagging via Particle Clouds

classification hep-ph cs.CVhep-ex
keywords particlecloudparticlenettaggingarchitecturecloudsjetsnetwork
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

How to represent a jet is at the core of machine learning on jet physics. Inspired by the notion of point clouds, we propose a new approach that considers a jet as an unordered set of its constituent particles, effectively a "particle cloud". Such a particle cloud representation of jets is efficient in incorporating raw information of jets and also explicitly respects the permutation symmetry. Based on the particle cloud representation, we propose ParticleNet, a customized neural network architecture using Dynamic Graph Convolutional Neural Network for jet tagging problems. The ParticleNet architecture achieves state-of-the-art performance on two representative jet tagging benchmarks and is improved significantly over existing methods.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 35 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Predict before you train: Scaling Laws for particle physics foundation models

    hep-ex 2026-07 conditional novelty 7.0

    A Chinchilla-style law fit on ParticleViT runs below 10^19 FLOPs predicts held-out pretraining loss within ~1% at >100× compute and tracks downstream jet-tagging rejection.

  2. Benchmarking Machine Learning Architectures for ttH Multilepton Signal Sensitivity

    hep-ph 2026-07 conditional novelty 7.0

    A controlled benchmark of six ML classifiers on a new simulated ttH multilepton dataset finds symmetry-constrained graph models (Particle Transformer, LorentzNet) and azimuthal RoPE encoding outperform tabular baselines.

  3. ParticleTransformer is all you need for reconstructing hadronic tau leptons

    hep-ex 2026-06 unverdicted novelty 7.0

    First fully machine-learned hadronic tau reconstruction at FCC-ee using ParticleTransformer achieves high performance on simulated data for identification, decay mode, charge, and kinematics.

  4. ParticleTransformer is all you need for reconstructing hadronic tau leptons

    hep-ex 2026-06 unverdicted novelty 7.0

    ParticleTransformer models achieve per-mille level misidentification rates, F1 scores up to 0.95 for decay modes, and superior kinematic resolution for hadronic tau leptons compared to conventional jet observables on ...

  5. Particle-Lund Multimodality in Jet Taggers

    hep-ph 2026-05 unverdicted novelty 7.0

    PLuM multimodal transformer improves top and H->bb jet tagging by jointly processing particle constituents and Lund plane splittings, yielding 25% higher background rejection at 25% di-Higgs efficiency.

  6. Higgs Physics with the XFEL Compton $\boldsymbol{\gamma\gamma}$ Collider Concept at $\boldsymbol{\sqrt{s}=125}$ GeV

    hep-ph 2026-05 unverdicted novelty 7.0

    An XFEL Compton gamma-gamma collider at 125 GeV with a set transformer deep learning classifier on particle-flow point clouds can achieve high-precision Higgs measurements across hadronic, semi-leptonic, and leptonic ...

  7. Lund Plane to Bloch (LP2B) Encoding for Object and Polarization Tagging with Quantum Jet Substructure

    quant-ph 2026-04 unverdicted novelty 7.0

    LP2B encoding converts Lund plane jet representations into Bloch sphere qubit states, enabling a QTTN that matches classical LundNet performance on polarization tagging and W/top tagging with three orders of magnitude...

  8. Probing Proton Structure via Physics-Guided Neural Networks in Holographic QCD

    hep-ph 2026-04 unverdicted novelty 7.0

    A physics-guided neural network embedding AdS5 Dirac equation and holographic Pomeron fits SLAC proton F2 data with chi-squared per degree of freedom of 0.91 and identifies a kinematic crossover at x approximately 0.1...

  9. Measurement of the jet mass in hadronic decays of boosted W bosons at 13 TeV and extraction of the W boson mass

    hep-ex 2026-03 accept novelty 7.0

    Unfolded double-differential W+jets cross section versus jet p_T and soft-drop mass yields m_W = 80.83 ± 0.55 GeV, the most precise all-jets extraction at a hadron collider.

  10. IAFormer: Interaction-Aware Transformer network for collider data analysis

    hep-ph 2025-05 unverdicted novelty 7.0

    IAFormer uses boost-invariant pairwise quantities and differential attention to create a sparse Transformer that achieves state-of-the-art classification on top-quark and quark-gluon jet datasets while using over an o...

  11. Search for dark matter in a signature with a four-prong large-radius jet in proton-proton collisions at $\sqrt{s}$ = 13 TeV

    hep-ex 2026-07 accept novelty 6.5

    No significant excess is observed in the four-prong large-radius jet + MET signature; 95% CL upper limits are set on the signal strength versus mediator mass or χ₂χ₁Y₀ coupling.

  12. Search for nonresonant triple Higgs boson production in the final state with six bottom quarks in proton-proton collisions at $\sqrt{s}$ = 13 TeV

    hep-ex 2026-07 accept novelty 6.0

    No excess is observed; the 95% CL upper limit on nonresonant HHH→6b is 44 fb (588×SM), with κ3 constrained to −7.4 < κ3 < 12.4 (κ4=1) and κ4 to −177 < κ4 < 185 (κ3=1).

  13. Exclusive Quark and Gluon Dijet Production as Probes of GPDs at Collider Energies

    hep-ph 2026-07 conditional novelty 6.0

    Exclusive dijet electroproduction probes GPDs via new LO amplitudes for quark (with helicity and QED) and gluon channels, with HERA comparison and EIC projections.

  14. Higher-order effects in amplitude-assisted polarisation extraction with machine-learning techniques

    hep-ph 2026-07 unverdicted novelty 6.0

    First NLO-QCD amplitude-assisted ML regression for longitudinal-boson production rate in di-boson events at the LHC, benchmarked against random forests.

  15. ParticleTransformer is all you need for reconstructing hadronic tau leptons

    hep-ex 2026-06 conditional novelty 6.0

    Transformer models trained on FCC-ee CLD simulation reconstruct hadronic taus with per-mille mis-ID, F1 up to 0.95, sub-per-mille charge errors, and percent-level transverse momentum resolution.

  16. Uncovering Hidden Systematics in Neural Network Models for High Energy Physics

    cs.LG 2026-05 unverdicted novelty 6.0

    Neural networks for HEP tasks can be fooled at significant rates by subtle perturbations inside uncertainty envelopes, revealing hidden systematics not captured by conventional methods.

  17. Search for pair production of heavy resonances in final states with a photon and large-radius jets in proton-proton collisions at $\sqrt{s}$ = 13 TeV

    hep-ex 2026-02 accept novelty 6.0

    Using 138 fb^-1 of CMS data, the first tt-gamma-g channel search finds no excess and excludes spin-1/2 (spin-3/2) excited top quarks below 930 (1330) GeV.

  18. Combination of searches for heavy vector boson resonances in proton-proton collisions at $\sqrt{s}$ = 13 TeV

    hep-ex 2026-01 conditional novelty 6.0

    A CMS combination of searches finds no heavy vector boson resonance and excludes HVT W′/Z′ bosons below 5.5 TeV (weak coupling), 4.8 TeV (strong coupling), and 2.0 TeV for VBF production at 95% CL.

  19. Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges

    cs.LG 2021-04 accept novelty 6.0

    Geometric deep learning provides a unified mathematical framework based on grids, groups, graphs, geodesics, and gauges to explain and extend neural network architectures by incorporating physical regularities.

  20. Exclusive Quark and Gluon Dijet Production as Probes of GPDs at Collider Energies

    hep-ph 2026-07 unverdicted novelty 5.0

    Exclusive quark and gluon dijet electroproduction is analyzed in collinear factorization as a GPD probe, extending prior work with helicity GPDs and an electromagnetic channel, with HERA comparisons and EIC projections.

  21. Statistical sensitivity of neutrinoless double-beta decay exchange mechanism discrimination by tracking experiments

    hep-ph 2026-05 unverdicted novelty 5.0

    Tracking detectors achieve 1σ mechanism discrimination in neutrinoless double-beta decay with a few events and 3σ with roughly 10 events, rising to about 25 events under realistic reconstruction uncertainties.

  22. Probing SMEFT Operators through $t\bar{t}t\bar{t}$ Production with Hyper-Graph Neural Networks at the LHC

    hep-ph 2026-05 unverdicted novelty 5.0

    A hyper-graph neural network improves discrimination of four-top production at 13 TeV, raising expected significance from 5.13 to 9.11 and enabling projected 95% CL limits on five dimension-six SMEFT Wilson coefficien...

  23. Explainable AI for Jet Tagging: A Comparative Study of GNNExplainer, GNNShap, and GradCAM for Jet Tagging in the Lund Jet Plane

    hep-ph 2026-04 unverdicted novelty 5.0

    Explainability techniques applied to LundNet show that assigned node importance correlates with classical jet substructure observables such as N-subjettiness ratios and energy correlation functions, with shifts across...

  24. Search for the pair production of long-lived supersymmetric partners of the tau lepton in proton-proton collisions at $\sqrt{s}$ = 13 TeV

    hep-ex 2026-01 accept novelty 5.0

    No long-lived staus are observed; CMS excludes stau masses up to 425 GeV at 50 mm lifetime (mass-degenerate) and decay lengths 21-94 mm / 6-333 mm at 200 GeV.

  25. Search for dark matter produced in association with a Higgs boson decaying to bottom quarks in proton-proton collisions at $\sqrt{s}$ = 13 TeV

    hep-ex 2026-01 accept novelty 5.0

    A CMS search with 138 fb^-1 of 13 TeV data finds no dark matter produced with a Higgs boson decaying to bottom quarks, and sets 95% CL exclusions on Z' and 2HDM+a model parameters.

  26. Search for a boosted Higgs boson decaying to bottom quark pairs in association with a W or Z boson in proton-proton collisions at $\sqrt{s}$ = 13 TeV

    hep-ex 2026-01 accept novelty 5.0

    In 138 fb^-1 of 13 TeV proton collisions, CMS measures the boosted VH, H->bb signal strength as mu=0.72^{+0.75}_{-0.71}, consistent with the Standard Model.

  27. Probing new physics in the Boosted $HH \to b\bar{b}\gamma\gamma$ channel at the LHC

    hep-ph 2025-12 conditional novelty 5.0

    A dedicated boosted-jet category in HH→bbγγ is projected to narrow the κ2V constraint to [-0.4, 2.6] at 95% CL and improve heavy-resonance limits by 1–2x at 308 fb⁻¹.

  28. KIGNet: Physics-Motivated Multi-Graph Representation Learning for Explainable Jet Tagging

    hep-ph 2025-12 conditional novelty 5.0

    E-PCN reaches 94.67% macro-accuracy on 10-class jet tagging by weighting graphs with angular separation, transverse momentum, momentum fraction, and invariant mass, with Grad-CAM showing the first two account for 76% ...

  29. KIGNet: Physics-Motivated Multi-Graph Representation Learning for Explainable Jet Tagging

    hep-ph 2025-12 conditional novelty 5.0

    An edge-weighted multi-graph GNN (E-PCN/KIGNet) that encodes four jet kinematic variables improves JetClass jet-tagging accuracy and attributes most predictions to angular separation and transverse momentum.

  30. Learning from all particles in high-energy collisions

    hep-ex 2025-06 unverdicted novelty 5.0

    Deep learning on all particles via holistic analysis and Advanced Color Singlet Identification improves Higgs signal extraction up to sixfold in high-energy collisions.

  31. Improving robustness of jet tagging algorithms with adversarial training: exploring the loss surface

    hep-ex 2023-03 unverdicted novelty 5.0

    Adversarial training enhances robustness of jet tagging classifiers while preserving performance, with loss surface geometry providing insights into correlations and vulnerability.

  32. The PYTHIA Facility

    hep-ph 2026-03 conditional novelty 4.0

    PYTHIA is presented as a 'big science facility' in software form: since 2018 its manuals drew ~9,600 citing works and ~47,000 unique authors across LHC, heavy-ion, flavor, astroparticle, and ML-for-physics communities.

  33. Projections of H$\to\tau\tau$ cross-section at FCC-ee

    hep-ph 2026-01 unverdicted novelty 4.0

    FCC-ee projections indicate at least 10 times better precision on the H to tau tau cross-section than the LHC through ZH and VBF channels plus improved tau reconstruction methods.

  34. HEPTAPOD: Orchestrating High Energy Physics Workflows Towards Autonomous Agency

    hep-ph 2025-12 conditional novelty 4.0

    HEPTAPOD uses LLM agents to drive FeynRules, MadGraph, Pythia, and analysis tools through schema-validated tool calls and run-card templates, demonstrated on a leptoquark signal scan.

  35. Looking inside jets: an introduction to jet substructure and boosted-object phenomenology

    hep-ph 2019-01