Cite
Notes
Only stored in your browser.
Attribution
Drop-Upcycling: Training Sparse Mixture of Experts with Partial Re-initialization
arXiv 2025
Rewriting Pre-Training Data Boosts LLM Performance in Math and Code
Heron-Bench: A Benchmark for Evaluating Vision Language Models in Japanese
arXiv 2024
from 3 papers
Rio Yokota
Taishi Nakamura
Hinari Shimada
Hiroya Takamura
Jun Suzuki
Kakeru Hattori
Kento Sasaki
Koshiro Saito
Kotaro Tanahashi
Masaki Kawamura