activity
20242026
collaborators

7 papers

cs.LG2026

Rewriting Pre-Training Data Boosts LLM Performance in Math and Code

Kazuki Fujii, Yukito Tajima, Sakae Mizuki +14

The performance of large language models (LLMs) in program synthesis and mathematical reasoning is fundamentally limited by the quality of their pre-training corpora. We introduce…

cs.CL2025

Why We Build Local Large Language Models: An Observational Analysis from 35 Japanese and Multilingual LLMs

Koshiro Saito, Sakae Mizuki, Masanari Ohi +11

Why do we build local large language models (LLMs)? What should a local LLM learn from the target language? Which abilities can be transferred from other languages? Do language-spe…

cs.CL2025

Building Instruction-Tuning Datasets from Human-Written Instructions with Open-Weight Large Language Models

Youmi Ma, Sakae Mizuki, Kazuki Fujii +12

Instruction tuning is crucial for enabling Large Language Models (LLMs) to solve real-world tasks. Prior work has shown the effectiveness of instruction-tuning data synthesized sol…

cs.CL2025

Drop-Upcycling: Training Sparse Mixture of Experts with Partial Re-initialization

Taishi Nakamura, Takuya Akiba, Kazuki Fujii +3

The Mixture of Experts (MoE) architecture reduces the training and inference cost significantly compared to a dense model of equivalent capacity. Upcycling is an approach that init…

cs.CL2024

LLM-jp: A Cross-organizational Project for the Research and Development of Fully Open Japanese LLMs

LLM-jp, :, Akiko Aizawa +80

This paper introduces LLM-jp, a cross-organizational project for the research and development of Japanese large language models (LLMs). LLM-jp aims to develop open-source and stron…

cs.LG2024

Balancing Speed and Stability: The Trade-offs of FP8 vs. BF16 Training in LLMs

Kazuki Fujii, Taishi Nakamura, Rio Yokota

Large Language Models (LLMs) have attracted significant attention due to their human-like language understanding and generation capabilities, as well as their applicability across…