REVIEW 11 cited by
Jet Diffusion versus JetGPT -- Modern Networks for the LHC
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
Signed reviews
read the original abstract
We introduce two diffusion models and an autoregressive transformer for LHC physics simulations. Bayesian versions allow us to control the networks and capture training uncertainties. After illustrating their different density estimation methods for simple toy models, we discuss their advantages for Z plus jets event generation. While diffusion networks excel through their precision, the transformer scales best with the phase space dimensionality. Given the different training and evaluation speed, we expect LHC physics to benefit from dedicated use cases for normalizing flows, diffusion models, and autoregressive transformers.
Forward citations
Cited by 11 Pith papers
-
Neural Control Variates at LO and NLO
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.
-
Learning Standard Model structure from LHC data with Riemannian flow matching
ShellFlow, a Riemannian flow-matching transformer fed only on-shell and invariant-mass priors and ~8×10^8 recorded ATLAS events, reproduces the SM's dilepton resonances, Weinberg angle, and top/W mass peaks in a singl...
-
Generative Amplification with Surrogate Monte Carlo
An amplitude surrogate trained on a few thousand exact LHC amplitude points statistically outperforms the training data, with largest amplification in sparsely populated kinematic tails of Z+g and Z+4g production.
-
Agentic Re-Casting using Agentic Re-Simulations
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.
-
Forecasting Generative Amplification
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.
-
Analysis-ready Generative Unfolding
Generative unfolding is extended to handle backgrounds, acceptance, and efficiency effects in an unbinned, iterative pipeline, demonstrated at percent-level accuracy on Gaussian and Z+jets simulations.
-
Simulation-Prior Independent Neural Unfolding Procedure
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.
-
Efficient many-jet event generation with Flow Matching
A Continuous Normalizing Flow trained with Flow Matching improves unweighting efficiency in high-multiplicity Drell-Yan and top-pair event generation by factors of 150 and 17 over Vegas.
-
How to Unfold Top Decays
A conditional flow-matching network with batch-level conditioning and multi-mass training unfolds top-decay kinematics and extracts the top mass with reduced model bias.
-
Generating particle physics Lagrangians with transformers
A BART transformer can generate gauge-invariant Lagrangians from field content with over 90% accuracy on in-distribution data, though its performance drops on realistic Standard Model benchmarks.
-
HEPTAPOD: Orchestrating High Energy Physics Workflows Towards Autonomous Agency
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.
Discussion (0). Continue with ORCID to comment.