Rio Yokota

Shanghai Institute for Science of Science — China

Physics and Astronomy · Physical Sciences

23h-index1.8kcitations200works1.22yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Electromagnetic Scattering and Analysis(61), Parallel Computing and Optimization Techniques(31), Matrix Theory and Algorithms(25), Electromagnetic Simulation and Numerical Methods(23), Advanced Neural Network Applications(20)

Publications200 total

PowerCLIP: Powerset Alignment for Contrastive Pre-Training
arXiv (Cornell University)·2025Open Access
FedPM: Federated Learning Using Second-order Optimization with Preconditioned Mixing of Local Parameters
arXiv (Cornell University)·2025Open Access
Masked Gated Linear Unit
arXiv (Cornell University)·2025Open Access
Improving LoRA with Variational Learning
arXiv (Cornell University)·2025Open Access
Variational Learning Finds Flatter Solutions at the Edge of Stability
arXiv (Cornell University)·2025Open Access
Scaling Large-scale GNN Training to Thousands of Processors on CPU-based Supercomputers
2025
Rewriting Pre-Training Data Boosts LLM Performance in Math and Code
arXiv (Cornell University)·2025Open Access
Quantum Turbulence Coupled with Externally Driven Normal-Fluid Turbulence in Counterflow of Superfluid <sup>4</sup>He
Journal of the Physical Society of Japan·2025
Continuously Visual Pre-training with Artificially Generated and Real Images
Journal of the Japan Society for Precision Engineering·2025Open Access
A General and Scalable GCN Training Framework on CPU Supercomputers
2025Open Access
Drop-Upcycling: Training Sparse Mixture of Experts with Partial Re-initialization
arXiv (Cornell University)·2025Open Access
Synthetic Dataset Pre-training for Precision Medical Segmentation Using Vision Transformers
Electronic Imaging·2025Open Access
Leveraging High-Resource English Corpora for Cross-lingual Domain Adaptation in Low-Resource Japanese Medicine via Continued Pre-training
2025Open Access
Why We Build Local Large Language Models: An Observational Analysis from 35 Japanese and Multilingual LLMs
arXiv (Cornell University)·2024· 1 citedOpen Access
On the Relationship Between Double Descent of CNNs and Shape/Texture Bias Under Learning Process
Lecture notes in computer science·2024Open Access
Scaling Large-scale GNN Training to Thousands of Processors on CPU-based Supercomputers
arXiv (Cornell University)·2024Open Access
Lion Cub: Minimizing Communication Overhead in Distributed Lion
arXiv (Cornell University)·2024Open Access
Scaling Backwards: Minimal Synthetic Pre-Training?
Lecture notes in computer science·2024· 1 cited
Rethinking Image Super-Resolution from Training Data Perspectives
Lecture notes in computer science·2024· 2 cited
Balancing Speed and Stability: The Trade-offs of FP8 vs. BF16 Training in LLMs
arXiv (Cornell University)·2024Open Access
page 1 of 10Next →

Frequent Co-authors

Kazuki Osawa(9), Lorena A. Barba(9), Yuichiro Ueno(7), David E. Keyes(7), Yohei Tsuji(6), Akira Naruse(6), Huda Ibeid(6), Kenji Yasuoka(4), Shinnosuke Obi(4), Hirokatsu Kataoka(4), Nakamasa Inoue(4), Tetsu Narumi(3), Hiroyuki Ootomo(3), Ryo Hayamizu(3), Sora Takashima(3), Satoshi Matsuoka(3), Ichitaro Yamazaki(3), Akihiro Ida(3), Tsuyoshi Hamada(2), Makoto Taiji(2)