Score: 1

PLaMo 2 Technical Report

Published: September 5, 2025 | arXiv ID: 2509.04897v2

By: Preferred Networks , : , Kaizaburo Chubachi and more

Potential Business Impact:

Makes computers understand Japanese much better.

Business Areas:
Natural Language Processing Artificial Intelligence, Data and Analytics, Software

In this report, we introduce PLaMo 2, a series of Japanese-focused large language models featuring a hybrid Samba-based architecture that transitions to full attention via continual pre-training to support 32K token contexts. Training leverages extensive synthetic corpora to overcome data scarcity, while computational efficiency is achieved through weight reuse and structured pruning. This efficient pruning methodology produces an 8B model that achieves performance comparable to our previous 100B model. Post-training further refines the models using a pipeline of supervised fine-tuning (SFT) and direct preference optimization (DPO), enhanced by synthetic Japanese instruction data and model merging techniques. Optimized for inference using vLLM and quantization with minimal accuracy loss, the PLaMo 2 models achieve state-of-the-art results on Japanese benchmarks, outperforming similarly-sized open models in instruction-following, language fluency, and Japanese-specific knowledge.

Page Count
29 pages

Category
Computer Science:
Computation and Language