Related papers: Reconstructing hadronically decaying tau leptons w…

Reconstructing Hamiltonians from local measurements is key to enabling reliable quantum simulation: both validating the implemented model, and identifying any left-over terms with sufficient precision is a problem of increasing importance.…

Pretrained models are the standard starting point for training. This approach consistently outperforms the use of a random initialization. However, pretraining is a costly endeavour that few can undertake. In this paper, we create better…

In general-purpose particle detectors, the particle-flow algorithm may be used to reconstruct a comprehensive particle-level view of the event by combining information from the calorimeters and the trackers, significantly improving the…

Deep learning models are often deployed in downstream tasks that the training procedure may not be aware of. For example, models solely trained to achieve accurate predictions may struggle to perform well on downstream tasks because…

Metal artifacts from high-attenuation implants severely degrade CT image quality, obscuring critical anatomical structures and posing a challenge for standard deep learning methods that require extensive paired training data. We propose a…

Existing deep learning methods for remote sensing image fusion often suffer from poor generalization when applied to unseen datasets due to the limited availability of real training data and the domain gap between different satellite…

Typically, loss functions, regularization mechanisms and other important aspects of training parametric models are chosen heuristically from a limited set of options. In this paper, we take the first step towards automating this process,…

In relativistic heavy-ion collisions, where quark-gluon plasma forms, hadron production is anisotropic at both low and high transverse momentum, driven by flow dynamics and spatial anisotropies. To better understand these mechanisms, we use…

Fine-tuning a pre-trained model on a downstream task often degrades its original capabilities, a phenomenon known as "catastrophic forgetting". This is especially an issue when one does not have access to the data and recipe used to develop…

New TeV-scale physics processes at the LHC can produce Higgs bosons with substantive transverse Lorentz boost, such that the Higgs's decay products are nominally contained in a single jet. In the case of a light Higgs decaying predominantly…

Boosted top quark tagging is one of the challenging, and at the same time exciting, tasks in high energy physics experiments, in particular in the exploration of new physics signals at the LHC. Several techniques have already been developed…

Photomultiplier tubes (PMTs) are widely used in particle and nuclear physics experiments. The reconstruction of PMT waveforms is a fundamental task in these experiments, where accurate extraction of photoelectron (PE) multiplicities and…

With the rise of powerful foundation models, a pre-training-fine-tuning paradigm becomes increasingly popular these days: A foundation model is pre-trained using a huge amount of data from various sources, and then the downstream users only…

Federated Parameter-Efficient Fine-Tuning (Fed-PEFT) enables lightweight adaptation of large pre-trained models in federated learning settings by updating only a small subset of parameters. However, Fed-PEFT methods typically assume a fixed…

The Phase-2 Upgrade of the CMS Level-1 Trigger (L1T) will reconstruct particles using the Particle Flow algorithm, connecting information from the tracker, muon, and calorimeter detectors, and enabling fine-grained reconstruction of high…

We explore the potential to use machine learning methods to search for heavy neutrinos, from their hadronic final states including a fat-jet signal, via the processes $pp \rightarrow W^{\pm *}\rightarrow \mu^{\pm} N \rightarrow \mu^{\pm}…

We introduce multiple physics pretraining (MPP), an autoregressive task-agnostic pretraining approach for physical surrogate modeling of spatiotemporal systems with transformers. In MPP, rather than training one model on a specific physical…

The rapid development of AI systems has been greatly influenced by the emergence of foundation models. A common approach for targeted problems involves fine-tuning these pre-trained foundation models for specific target tasks, resulting in…

Large pre-trained models, or foundation models, have shown impressive performance when adapted to a variety of downstream tasks, often out-performing specialized models. Hypernetworks, neural networks that generate some or all of the…

Advancements in machine learning for molecular property prediction have improved accuracy but at the expense of higher computational cost and longer training times. Recently, the Joint Multi-domain Pre-training (JMP) foundation model has…