← Search

Kazuki Yano

2 accepted papers

2026

Pre-training LLM without Learning Rate Decay Enhances Supervised Fine-Tuning

ICLR 2026poster

We investigate the role of learning rate scheduling in the large-scale pre-training of large language models, focusing on its influence on downstream performance after supervised fine-tuning (SFT). Decay-based learning rate schedulers are widely used to minimize pre-training loss. However, despite t…

Cited by 0SourcecodeScholar
2025

STEP: Staged Parameter-Efficient Pre-training for Large Language Models

NAACL 2025short

Pre-training large language models (LLMs) faces significant memory challenges due to the large size of model weights. We introduce STaged parameter-Efficient Pre-training (STEP), which integrates parameter-efficient tuning techniques with model growth. We conduct experiments on pre-training LLMs of…