Methods and Stability Tests Associated with the Sterile Neutrino Search Using Improved High-Energy Ν_μ Event Reconstruction in IceCube
PHYSICAL REVIEW D(2024)
University of Canterbury Dept. of Physics and Astronomy | University of Wisconsin-Madison Dept. of Physics and Wisconsin IceCube Particle Astrophysics Center | Université Libre de Bruxelles | Niels Bohr Institute | TU Dortmund University Dept. of Physics | University of Delaware Bartol Research Institute and Dept. of Physics and Astronomy | Marquette University Department of Physics | Harvard University Department of Physics | RWTH Aachen University III. Physikalisches Institut | University of California Dept. of Physics and Astronomy | Technische Universität München Physik-department | University of California Dept. of Physics | Ohio State University Dept. of Physics and Center for Cosmology and Astro-Particle Physics | Loyola University Chicago Department of Physics | Fakultät für Physik & Astronomie | University of Maryland Dept. of Physics | University of Kansas Dept. of Physics and Astronomy | Karlsruhe Institute of Technology Institute for Astroparticle Physics | University of Mainz Institute of Physics | Georgia Institute of Technology School of Physics and Center for Relativistic Astrophysics | University of Adelaide Department of Physics | University of Nevada Nevada Center for Astrophysics | Massachusetts Institute of Technology Dept. of Physics | Vrije Universiteit Brussel (VUB) | Université catholique de Louvain Centre for Cosmology | Westfälische Wilhelms-Universität Münster Institut für Kernphysik | Erlangen Centre for Astroparticle Physics | Karlsruhe Institute of Technology Institute of Experimental Particle Physics | Chiba University Dept. of Physics | Southern University Dept. of Physics | Institut für Physik | University of Wisconsin-Madison Dept. of Astronomy | Lawrence Berkeley National Laboratory | Michigan State University Dept. of Physics and Astronomy | Queen's University Engineering Physics | Clark-Atlanta University CTSPS | University of Texas at Arlington Dept. of Physics | University of Alberta Dept. of Physics | Lawrence Berkeley National Laboratory Dept. of Physics | Université de Genève | Yale University Dept. of Physics | Mercer University Department of Physics | University of Gent Dept. of Physics and Astronomy | University of Alaska Anchorage Dept. of Physics and Astronomy
- Pretraining has recently greatly promoted the development of natural language processing (NLP)
- We show that M6 outperforms the baselines in multimodal downstream tasks, and the large M6 with 10 parameters can reach a better performance
- We propose a method called M6 that is able to process information of multiple modalities and perform both single-modal and cross-modal understanding and generation
- The model is scaled to large model with 10 billion parameters with sophisticated deployment, and the 10 -parameter M6-large is the largest pretrained model in Chinese
- Experimental results show that our proposed M6 outperforms the baseline in a number of downstream tasks concerning both single modality and multiple modalities We will continue the pretraining of extremely large models by increasing data to explore the limit of its performance
