Pith. sign in

REVIEW 10 cited by

OmniJet-$\alpha$: The first cross-task foundation model for particle physics

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 2403.05618 v2 pith:MSNXOL5B submitted 2024-03-08 hep-ph cs.LGhep-exphysics.data-an

classification hep-phcs.LGhep-exphysics.data-an
keywords physicsfoundationmodelsdatafirstparticlealphageneration
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

Foundation models are multi-dataset and multi-task machine learning methods that once pre-trained can be fine-tuned for a large variety of downstream applications. The successful development of such general-purpose models for physics data would be a major breakthrough as they could improve the achievable physics performance while at the same time drastically reduce the required amount of training time and data. We report significant progress on this challenge on several fronts. First, a comprehensive set of evaluation methods is introduced to judge the quality of an encoding from physics data into a representation suitable for the autoregressive generation of particle jets with transformer architectures (the common backbone of foundation models). These measures motivate the choice of a higher-fidelity tokenization compared to previous works. Finally, we demonstrate transfer learning between an unsupervised problem (jet generation) and a classic supervised task (jet tagging) with our new OmniJet-$\alpha$ model. This is the first successful transfer between two different and actively studied classes of tasks and constitutes a major step in the building of foundation models for particle physics.

Discussion (0). Sign in to comment.

Forward citations

Cited by 10 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 of 10

    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. Learning Standard Model structure from LHC data with Riemannian flow matching

    hep-ph 2026-07 conditional novelty 7.0 of 10

    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...

  3. Neural Scaling Laws for Jet Generation

    hep-ph 2026-05 unverdicted novelty 7.0 of 10

    Scaling laws hold logarithmically for model size in autoregressive jet generation, with next-token loss correlating to physical metrics via sliced Wasserstein distance, but show weaker scaling for dataset size and com...

  4. Learning transferable event representations for charmed baryon physics at BESIII

    physics.data-an 2026-07 conditional novelty 6.0 of 10

    A Particle Transformer pre-trained on simulated Lambda_c events transfers across 12 decay channels, improving classification and momentum-direction regression over training from scratch in low-statistics regimes.

  5. SPADE: Split-and-Delay Embeddings for Autoregressive High-Granularity Calorimeter Simulation

    physics.ins-det 2026-06 unverdicted novelty 6.0 of 10

    SPADE is a split-and-delay embedding technique for multi-feature autoregressive transformers that achieves competitive performance on high-granularity calorimeter shower simulation.

  6. Explicit or Implicit? Encoding Physics at the Precision Frontier

    hep-ph 2026-03 conditional novelty 6.0 of 10

    On three precision classification tasks — reweighting-based unfolding, likelihood-ratio estimation, and weakly supervised anomaly detection — a Lorentz-equivariant transformer and a pretrained foundation model perform...

  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. Enhancing next token prediction based pre-training for jet foundation models

    hep-ph 2025-12 conditional novelty 6.0 of 10

    Using continuous particle features as input and combining next-token with masked-token pre-training markedly improves classification accuracy of the OmniJet jet foundation model without visibly hurting its generative quality.

  9. Are We Ready for AI-Driven Discovery? AI Verification Before the Next Fundamental Physics Breakthrough

    physics.data-an 2026-07 accept novelty 4.0 of 10

    Verification of ML in fundamental physics is essential precisely when models enter statistical modeling, inference, or hypothesis testing, and is bounded by unavoidable inductive bias, sample complexity, and experimen...

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

    hep-ph 2025-12 conditional novelty 4.0 of 10

    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.

Pith tools