The paper introduces a Deep Sets surrogate for optimal transport (OT) that respects key metric properties—non-negativity, exchange symmetry, and zero self-distance—while leaving the triangle inequality unconstrained. Applied to the Energy Mover's Distance between collider events, the Metric-Aware Particle Flow Network achieves percent‑level mean absolute percentage error and markedly higher inference throughput compared to other exact and approximate methods. The architectural constraints also dramatically reduce triangle‑inequality violations, improving geometric fidelity across a large set of held‑out event triplets.
By Lauren Hay, Rishabh Jain, Matt LeBlanc, Jennifer Roloff
VyPER is a new geometric learning framework that reconstructs particle collider events by representing them as hypergraphs with a physics-inspired topology. It tackles two key tasks: assigning measured jets and leptons to their parent particles through supervised hyperedge classification, and predicting neutrino kinematics using a diffusion model, all optimized jointly with a shared loss function. The authors evaluate VyPER on various proton‑proton collision processes, showing improved performance over existing analytical and machine‑learning methods and enabling more precise measurements in Higgs, electroweak, and top‑quark studies.
By Lining Mao, Yvonne Peters, Ethan Simpson, Zihan Zhang
arXiv:2606. 14813v1 Announce Type: cross Abstract: Jet tagging at the Large Hadron Collider increasingly relies on deep learning models trained on massive simulated datasets, leading to high computational costs and limited robustness to detector mismodeling.
By Guillaume Letellier (LPCC), Antonin Vacheret (LPCC), Fr\'ed\'eric Jurie
arXiv:2606. 14373v1 Announce Type: cross Abstract: The workflow from particle collision to physics analysis passes through a series of reconstruction steps that are traditionally modular and disconnected, with no shared representation linking low-level detector data to high-level analysis tasks.
By Farouk Mokhtar, Joosep Pata, Michael Kagan, Javier Duarte
The paper introduces HyPER, a hypergraph-based graph neural network architecture designed to reconstruct short-lived particles in collider experiments. By leveraging hypergraph representation learning, HyPER builds more powerful and efficient representations of collider events, enabling accurate reconstruction of parent particles from final-state objects. In simulations, HyPER outperforms existing state‑of‑the‑art techniques while using fewer parameters, and its flexible hypergraph approach can be applied to a wide range of physics processes.
By Callum Birch-Sykes, Brian Le, Yvonne Peters, Ethan Simpson, Zihan Zhang
Panda Diplomacy introduces a point‑cloud self‑distillation framework that enables a single foundation‑model architecture and objective to be pre‑trained across three distinct particle‑detector modalities—liquid argon time‑projection chambers, collider TPCs, and water Cherenkov detectors—without extensive modification. Using only 1,000 labeled images for downstream adaptation, the resulting Panda V2 model matches or surpasses specialized baselines that require orders of magnitude more supervision, achieving state‑of‑the‑art particle‑clustering performance with 70× fewer labeled events on sPHENIX and up to 1,000× fewer labels on LArTPC data. Linear probes further demonstrate that the model’s latent space captures physically meaningful structures such as particle causality and track curvature.
By Samuel Young, C\'esar Jes\'us-Valls, Kazuhiro Terao
The paper introduces ETHEREAL, the first accelerator for event-driven graph neural networks (EV‑GNNs) that can handle 640×480 resolution inputs. It achieves this through a neighbor‑parallel spline convolution engine and a 2D/3D‑split memory hierarchy that includes a novel region‑of‑interest spatiotemporal caching mechanism. Measurements show end‑to‑end inference latency of 25.6 µs and energy consumption of 1.7 µJ per event on state‑of‑the‑art workloads.
By Adrian Kneip, Martin Lefebvre, Daniel Gehrig, Victoria Catal\'an Pastor, Davide Scaramuzza, Marian Verhelst, Charlotte Frenkel
arXiv:2511. 22246v2 Announce Type: replace-cross Abstract: Unsupervised learning has been widely applied to various tasks in particle physics.
By Xing-Jian Lv, De-Xing Miao, Zi-Jun Xu, Jian-Chun Wang
arXiv:2608.24365v1 Announce Type: new
Abstract: Transformer-based object trackers are renowned for their strong performance, yet dense token processing often leads to prohibitive computational cost,...
By Qingmao Wei, Fagui Liu, Dengke Zhang, Qingze He, Quan Tang
TrackEverything is a 3D point tracker that overcomes the trade‑off between sparse long‑horizon tracking and dense short‑clip tracking by representing videos as persistent 3D scene tracks in world coordinates. It introduces voxel‑based de‑duplication at sliding‑window boundaries, a two‑stage refinement process (endpoint refiner and lightweight trajectory refiner), and a 3D WAFT module that replaces memory‑heavy 4D correlation volumes with efficient feature sampling. The method can track all visible points in videos longer than 1000 frames using only 40 GB of GPU memory, outperforming existing dense trackers on short clips and matching sparse trackers on long sequences.
By Ayush Jain, Sreeharsha Paruchuri, Ishita Gupta, Fan Zhang, Tanner Schmidt, Jakob Engel, Katerina Fragkiadaki, Adam W. Harley
arXiv:2407.04308v4 Announce Type: replace-cross
Abstract: We propose a graph-based tracking formulation for multi-object tracking (MOT) where target detections contain kinematic information and re-id...
By Griffin Golias, Masa Nakura-Fan, Vitaly Ablavsky
Transformer-based object trackers are renowned for their strong performance, yet dense token processing often leads to prohibitive computational cost, limiting real-time deployment on edge devices. Wh...