Pith. sign in

REVIEW 14 cited by

MadNIS -- Neural Multi-Channel Importance Sampling

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 2212.06172 v2 pith:TNXKNLKT submitted 2022-12-12 hep-ph hep-exphysics.comp-ph

classification hep-phhep-exphysics.comp-ph
keywords importanceintegrationmulti-channelnumericalsamplingadditionalbi-directionalbuffered
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Theory predictions for the LHC require precise numerical phase-space integration and generation of unweighted events. We combine machine-learned multi-channel weights with a normalizing flow for importance sampling, to improve classical methods for numerical integration. We develop an efficient bi-directional setup based on an invertible network, combining online and buffered training for potentially expensive integrands. We illustrate our method for the Drell-Yan process with an additional narrow resonance.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 14 Pith papers

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

  1. Neural Control Variates at LO and NLO

    hep-ph 2026-07 accept novelty 7.0 of 10

    Signed neural control variates from normalizing flows, combined with neural importance sampling, reduce weight ranges and negative weights for LO and NLO phase-space integration and event generation.

  2. Schr\"{o}dinger Generator for High-Dimensional Integration and Sampling on Quantum Many-Body States

    nucl-th 2026-08 reject novelty 6.0 of 10

    A two-stage sampler (adaptive marginal map plus normalizing flow, then resampling) is proposed and shown on model nuclear densities up to D=624, though a core Jacobian equation appears sign-inconsistent.

  3. Automated NRQCD and NRQED simulations of quarkonium and leptonium production with P-wave states and physical-mass effects

    hep-ph 2026-07 conditional novelty 6.0 of 10

    MadSONS extends MadGraph to automated LO NRQCD/NRQED event generation for arbitrary S- and P-wave bound states, with dual-number projectors and physical-mass reshuffling.

  4. Agentic Re-Casting using Agentic Re-Simulations

    hep-ph 2026-07 conditional novelty 6.0 of 10

    An agentic AI system with a physicist in the loop re-casts an ATLAS ttZ measurement into a global top-quark SMEFT fit and recovers injected coloron Wilson coefficients in a repeatable benchmark.

  5. An Optimal Transportation Approach for Improved Confidence Intervals

    stat.ME 2026-06 unverdicted novelty 6.0 of 10

    Optimal-transport couplings are used to construct confidence intervals that reduce coverage error relative to classical quantile-based intervals, with consistency theory and data-driven hyperparameters.

  6. CaloTrilogy: Toward a Breakthrough in One-Step, End-to-End, Physics-Guided Shower Generation for Modern Calorimeters

    hep-ex 2026-06 unverdicted novelty 6.0 of 10

    Presents CaloTrilogy, a unified one-step generative model for high-granularity calorimeter showers that combines velocity field integration, learned priors, and physics losses to match SOTA quality.

  7. A universal vision transformer for fast calorimeter simulations

    hep-ph 2026-01 conditional novelty 6.0 of 10

    A vision-transformer flow-matching model generates calorimeter showers across regular and irregular detector geometries at millisecond speeds, and pretraining plus fine-tuning cuts training cost by about half.

  8. Forecasting Generative Amplification

    hep-ph 2025-09 conditional novelty 6.0 of 10

    A KS-test-based differential method and a Bayesian averaging method estimate generative amplification without holdout datasets and find amplification in selected LHC phase-space regions.

  9. Data-parallel leading-order event generation in MadGraph5_aMC@NLO

    hep-ph 2025-07 conditional novelty 6.0 of 10

    CUDACPP gives MadGraph data-parallel helicity amplitudes, delivering linear SIMD CPU speed-ups and up to order-of-magnitude GPU speed-ups for high-multiplicity QCD event generation.

  10. Simulation-Prior Independent Neural Unfolding Procedure

    hep-ph 2025-07 conditional novelty 6.0 of 10

    SPINUP is a neural-unfolding method that fits a parton-level generative model directly to detector-level data through a learned forward simulator, aiming to remove the simulation-prior bias.

  11. A Demonstration of ARCANE Reweighting: Reducing the Sign Problem in the MC@NLO Generation of $e^+ e^- \rightarrow q \bar{q} + 1\, jet$ Events

    hep-ph 2025-02 conditional novelty 6.0 of 10

    ARCANE reweighting cuts the post-unweighting negative-event fraction in e+e- -> q qbar + 1 jet MC@NLO generation from about 2.25% to below 10^-5 while preserving the visible event distributions.

  12. ARCANE Reweighting: A Monte Carlo Technique to Tackle the Negative Weights Problem in Collider Event Generation

    hep-ph 2025-02 conditional novelty 6.0 of 10

    ARCANE reweighting adds a carefully designed, zero-average correction to event weights so that positive and negative pathways to the same event cancel, preserving all physical distributions.

  13. Toward an event-level analysis of hadron structure using differential programming

    hep-ph 2025-07 conditional novelty 4.0 of 10

    LOITS is a differentiable sampling method, demonstrated in a GAN closure test, that maps sampled events back to the parameters of a target density for event-level inference.

  14. Communicating Likelihoods with Normalising Flows

    hep-ph 2025-02 conditional novelty 4.0 of 10

    A normalizing-flow workflow compresses sample-based likelihoods into small files, validated with a radial Kolmogorov-Smirnov test on three high-energy physics examples.

Pith tools