Particle-Lund Multimodality in Jet Taggers

• URL: http://arxiv.org/abs/2605.26821v1
• Date: Tue, 26 May 2026 10:39:34 GMT
• Title: Particle-Lund Multimodality in Jet Taggers
• Abstract summary: Lund plane offers a physics-motivated, hierarchical representation of QCD radiation within jets.<n>Transformers have reached state-of-the-art performance by learning directly from raw particle constituents.<n>We investigate whether transformers implicitly capture hierarchical QCD structure from constituent-level inputs.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: The Lund plane offers a physics-motivated, hierarchical representation of QCD radiation within jets, while transformer-based taggers have reached state-of-the-art performance by learning directly from raw particle constituents and their pairwise relations. We investigate whether transformers implicitly capture hierarchical QCD structure from constituent-level inputs, or whether explicit physics representations remain complementary. To test this, we introduce PLuM, a multimodal architecture that projects particle constituents and Lund plane splittings into a shared latent space, processing both jointly with a unified transformer. Cross-attention allows the model to probe whether structured QCD information provides discriminating power beyond what particles alone encode. We observe systematic gains for top-quark and $\mathrm{H}\to\mathrm{b}\bar{\mathrm{b}}$ tagging, while finding no comparable improvement for $\mathrm{H}\to\mathrm{c}\bar{\mathrm{c}}$ or $\mathrm{H}\to 4\mathrm{q}$ topologies. This selective enhancement suggests that explicit hierarchical information about b-jet formation remains complementary to raw particle representations even in highly expressive architectures, while other topologies are already well-captured at constituent level. For high-impact LHC analyses such as Lorentz-boosted di-Higgs searches in the four $\mathrm{b}$ quark final state ($\mathrm{H}\mathrm{H}(4\mathrm{b})$), the gains are substantial: at a $25\%$ di-Higgs efficiency working point, PLuM achieves $25\%$ higher background rejection than the baseline. Our results indicate that physically structured representations of QCD radiation retain discriminating value in the transformer era, motivating further study into how different aspects of jet dynamics are encoded by deep learning algorithms.

Related papers

• An All-van-der-Waals Qubit [2.307942733255708]
  We report an all-vdW superconducting qubit based on a NbSe$$-hBN-NbSe$ junction.<n>Results show that lumped-element superconducting qubits can achieve coherence times comparable to those of conventional Al-AlO$_mathrmx$-Al qubits.
  arXiv  Detail & Related papers  (2026-09-13T01:46:57Z)
• WorldParticle: Unified World Simulation of Lagrangian Particle Dynamics via Transformer [107.36949516442782]
  We present a learning-based particle simulator built on a single transformer architecture to model cloth, elastic solds, Newtonian and non-Newtonian fluids, granular materials, and molecular dynamics.
  arXiv  Detail & Related papers  (2026-05-14T18:18:12Z)
• Block Permutation Routing on Ramanujan Hypergraphs for Fault-Tolerant Quantum Computing [0.0]
  We analyze permutation routing of rigid blocks representing surface code patches of $d_C2$ atoms on a reconfigurable lattice.<n>A spectral analysis of the quotient graph $Q(G_mathrmcl(H), B)$ shows that the spectral ratio $_Q 1$ is preserved in the high-connectivity regime.
  arXiv  Detail & Related papers  (2026-05-06T15:29:08Z)
• Gradient-Based Excitation Filter for Molecular Ground-State Simulation [6.627541720714792]
  We introduce a method to efficiently simplify the Unitary Coupled-Cluster with Single and Double Excitations (UCCSD) ansatz on classical computers.<n>We demonstrate that our approach achieves up to $46%$ parameter decrease, $60%$ circuit depth reduction and $678times runtime.
  arXiv  Detail & Related papers  (2025-06-25T13:12:45Z)
• UniF$^2$ace: A Unified Fine-grained Face Understanding and Generation Model [62.66515621965686]
  We introduce a novel theoretical framework with a Dual Discrete Diffusion (D3Diff) loss, unifying masked generative models with discrete score matching diffusion.<n>This D3Diff significantly enhances the model's ability to synthesize high-fidelity facial details aligned with text input.<n>We construct UniF$2$aceD-1M, a large-scale dataset comprising 130K fine-grained image-caption pairs and 1M visual question-answering pairs.
  arXiv  Detail & Related papers  (2025-03-11T07:34:59Z)
• Unveiling Induction Heads: Provable Training Dynamics and Feature Learning in Transformers [54.20763128054692]
  We study how a two-attention-layer transformer is trained to perform ICL on $n$-gram Markov chain data. We prove that the gradient flow with respect to a cross-entropy ICL loss converges to a limiting model.
  arXiv  Detail & Related papers  (2024-09-09T18:10:26Z)
• Supervised binary classification of small-scale digit images and weighted graphs with a trapped-ion quantum processor [56.089799129458875]
  We present the results of benchmarking a quantum processor based on trapped $171$Yb$+$ ions.<n>We perform a supervised binary classification on two types of datasets: small binary digit images and weighted graphs with a ring topology.
  arXiv  Detail & Related papers  (2024-06-17T18:20:51Z)
• A case study of sending graph neural networks back to the test bench for applications in high-energy particle physics [0.0]
  In high-energy particle collisions the primary collision products usually decay further resulting in tree-like, hierarchical structures with a priori unknown multiplicity. The analogy to mathematical graphs gives rise to the idea that graph neural networks (GNNs) should be best-suited to address many tasks related to high-energy particle physics. We describe a benchmark test of a typical GNN against neural networks of the well-established deep fully-connected feed-forward architecture.
  arXiv  Detail & Related papers  (2024-02-27T10:26:25Z)
• Explainable Equivariant Neural Networks for Particle Physics: PELICAN [51.02649432050852]
  PELICAN is a novel permutation equivariant and Lorentz invariant aggregator network. We present a study of the PELICAN algorithm architecture in the context of both tagging (classification) and reconstructing (regression) Lorentz-boosted top quarks. We extend the application of PELICAN to the tasks of identifying quark-initiated vs.gluon-initiated jets, and a multi-class identification across five separate target categories of jets.
  arXiv  Detail & Related papers  (2023-07-31T09:08:40Z)
• Rethinking SO(3)-equivariance with Bilinear Tensor Networks [0.0]
  We show that by judicious symmetry breaking, we can efficiently increase the expressiveness of a network operating only on vector and order-2 tensor representations of SO$(2)$. We demonstrate the method on an important problem from High Energy Physics known as textitb-tagging, where particle jets originating from b-meson decays must be discriminated from an overwhelming QCD background.
  arXiv  Detail & Related papers  (2023-03-20T17:23:15Z)
• Statistical Learning under Heterogeneous Distribution Shift [71.8393170225794]
  Ground-truth predictor is additive $mathbbE[mathbfz mid mathbfx,mathbfy] = f_star(mathbfx) +g_star(mathbfy)$.
  arXiv  Detail & Related papers  (2023-02-27T16:34:21Z)
• Clustering of Electromagnetic Showers and Particle Interactions with Graph Neural Networks in Liquid Argon Time Projection Chambers Data [4.653747487703939]
  Liquid Argon Time Projection Chambers (LArTPCs) are a class of detectors that produce high resolution images of charged particles within their sensitive volume. The clustering of distinct particles into superstructures is of central importance to the current and future neutrino physics program. This paper uses Graph Neural Networks (GNNs) to predict the adjacency matrix of EM shower fragments and to identify the origin of showers.
  arXiv  Detail & Related papers  (2020-07-02T18:32:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.