aiqbench analysis
This research introduces a data-driven method using energy mover's distance (EMD) to pair particle collision events for self-supervised pre-training, eliminating the need for computationally expensive or physics-breaking data augmentations. By leveraging similarity-based pairing instead of handcrafted distortions, the approach achieves semantic jet embeddings with discrimination power comparable to standard augmentation baselines while preserving event fidelity.
Similarity Pairing with Energy Mover's Distance for Self-Supervised Pre-Training at the LHC
arXiv:2609.17738v1 Announce Type: cross Abstract: Many self-supervised methods for training foundation models at the Large Hadron Collider (LHC) rely on data augmentations to encourage the model to embed events into a representation space invariant to certain physical or detector symmetries. A common challenge arises from the large freedom in choosing a proper set of augmentations on which downstream performance depends. The implementation of augmentations involves either modifying existing events, potentially breaking the event fidelity, or simulating more event variants, which is computationally intensive. In this work, we present a data-driven method of pairing events by their similarity via the energy mover's distance (EMD), which measures how similar two events are in terms of the work required to transform one into the other. With this approach, distinct events are sampled and matched by their similarity to serve as views for learning invariance, keeping the physics content of each event intact without handcrafted distortions. We demonstrate this augmentation-free pairing method by pre-training on QCD jets via self-distillation and show that it can yield semantic jet embeddings with downstream discrimination power comparable to or better than an augmentation-based baseline.
Abstract:Many self-supervised methods for training foundation models at the Large Hadron Collider (LHC) rely on data augmentations to encourage the model to embed events into a representation space invariant to certain physical or detector symmetries. A common challenge arises from the large freedom in choosing a proper set of augmentations on which downstream performance depends. The implementation of augmentations involves either modifying existing events, potentially breaking the event fidelity, or simulating more event variants, which is computationally intensive. In this work, we present a data-driven method of pairing events by their similarity via the energy mover's distance (EMD), which measures how similar two events are in terms of the work required to transform one into the other. With this approach, distinct events are sampled and matched by their similarity to serve as views for learning invariance, keeping the physics content of each event intact without handcrafted distortions. We demonstrate this augmentation-free pairing method by pre-training on QCD jets via self-distillation and show that it can yield semantic jet embeddings with downstream discrimination power comparable to or better than an augmentation-based baseline.
High Energy Physics - Experiment (hep-ex); Machine Learning (cs.LG)
arXiv:2609.17738 [hep-ex]
(or arXiv:2609.17738v1 [hep-ex] for this version)
https://doi.org/10.48550/arXiv.2609.17738 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 281 words
Related coverage
Import AI 476: Creativity and RSI; worse-than-China open weight models; AI goes to space
Import AI · Oct 12, 2026
The Continual Constitutional Convention: How language models revise the documents that govern them
AI Alignment Forum · Oct 13, 2026
MoCo-AIS: A Contrastive Learning Framework for Similarity Computation of Vessel Trajectories
arXiv · Jun 17, 2026
MoCo-AIS: A Contrastive Learning Framework for Similarity Computation of Vessel Trajectories
arXiv · Aug 25, 2026