Related papers: Learning PDFs through Interpretable Latent Represe…
Document parsing (DP) transforms unstructured or semi-structured documents into structured, machine-readable representations, enabling downstream applications such as knowledge base construction and retrieval-augmented generation (RAG).…
Machine learning has become a promising approach for molecular modeling. Positional quantities, such as interatomic distances and bond angles, play a crucial role in molecule physics. The existing works rely on careful manual design of…
Machine learning (ML) tools such as encoder-decoder convolutional neural networks (CNN) can represent incredibly complex nonlinear functions which map between combinations of images and scalars. For example, CNNs can be used to map…
Machine learning (ML) models are often constrained by their limitations in extrapolation, which restricts their applicability in engineering contexts. Conversely, while exhibiting broad generality, many established scientific models seem to…
Parton distributions are an essential ingredient of the LHC program. PDFs are relevant for precision Standard Model measurements, for Higgs boson characterization as well as for New Physics searches. In this contribution I review recent…
The recovery of training data from generative models ("model inversion") has been extensively studied for diffusion models in the data domain as a memorization/overfitting phenomenon. Latent diffusion models (LDMs), which operate on the…
This report summarizes the latest developments in the CTEQ-TEA global analysis of parton distribution functions (PDFs) in the nucleon. The focus is on recent NNLO fits to high-precision LHC data at 8 and 13 TeV, including Drell-Yan, jet,…
Nuclear parton distribution functions (nuclear PDFs) are non-perturbative objects that encode the partonic behaviour of bound nucleons. To avoid potential higher-twist contributions, the data probing the high-$x$ end of nuclear PDFs are…
The precise determination of the parton distribution functions (PDFs) of the proton is an essential ingredient for LHC analyses, including for those at the upcoming High-Luminosity LHC. So far, PDFs are determined from global fits to binned…
Decomposing a deep neural network's learned representations into interpretable features could greatly enhance its safety and reliability. To better understand features, we adopt a geometric perspective, viewing them as a learned coordinate…
Probability theory has become the predominant framework for quantifying uncertainty across scientific and engineering disciplines, with a particular focus on measurement and control systems. However, the widespread reliance on simple…
Current mainstream of CT reconstruction methods based on deep learning usually needs to fix the scanning geometry and dose level, which will significantly aggravate the training cost and need more training data for clinical application. In…
While protein language models (PLMs) are one of the most promising avenues of research for future de novo protein design, the way in which they transform sequences to hidden representations, as well as the information encoded in such…
We provide an assessment of the state of the art in various issues related to experimental measurements, phenomenological methods and theoretical results relevant for the determination of parton distribution functions (PDFs) and their…
We present Fanto10, a new ensemble of NLO error parton distribution functions (PDFs) in a charged pion that provides the most detailed estimate of uncertainties from experimental, theoretical, and methodological sources in order to enable…
We extract two nonsinglet nucleon Parton Distribution Functions from lattice QCD data for reduced Ioffe-time pseudodistributions. We perform such analysis within the NNPDF framework, considering data coming from different lattice ensembles…
Parton distribution function (PDF) at small $x$ in a fast-moving proton is investigated within an upgraded parton model that includes parton splitting with branching cascades and parton fusion. In the region of moderately small $x$, we…
Neutrino Deep Inelastic Scattering on nuclei is an essential process to constrain the strange quark parton distribution functions in the proton. The critical component on the way to using the neutrino DIS data in a proton PDF analysis is…
Large language models (LLMs) demonstrate remarkable emergent abilities to perform in-context learning across various tasks, including time series forecasting. This work investigates LLMs' ability to estimate probability density functions…
Quasi parton distribution functions (QPDFs) are defined in terms of QCD fields at spacelike separations evaluated in matrix elements of hadrons moving with velocity $v$. These objects can be studied in lattice QCD. In the limit when $v$…