English(EN) LLM-jp-4 Technical Report

LLM-jp-4 系列日语大模型发布，注重透明度

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-15 04:00

推出了一系列新的日语大语言模型 LLM-jp-4，其中包括一个 8B 的密集模型和一个 32B 的混合专家模型。这些模型在约 12 万亿个 token 上进行训练，注重透明度和可复现性。评估结果显示，与之前的 LLM-jp 版本相比有显著改进，并且与同等规模的开放权重模型相比具有竞争力。 AI

影响为未来具有透明度和可复现性的大模型训练的主权人工智能发展提供了参考。

排序理由该集群描述了一份技术报告，详细介绍了新一代大语言模型，包括其架构、训练数据和评估结果。[lever_c_demoted from research: ic=1 ai=1.0]

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 12 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群描述了一份技术报告，详细介绍了新一代大语言模型，包括其架构、训练数据和评估结果。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: model release, paper; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

arXiv cs.CL TIER_1 English(EN) · LLM-jp, :, Chaoran Liu, Daisuke Kawahara, Hideyuki Tachibana, Hirokazu Kiyomaru, Jun Suzuki, Koshi Eguchi, Kouta Nakayama, Qianying Liu, Rio Yokota, Ryo Takahashi, Sadao Kurohashi, Sosuke Hosokawa, Taishi Nakamura, Takashi Kodama, Yun-Ang Wu, Yusuke Miy… · 2026-10-15 04:00

LLM-jp-4 技术报告

arXiv:2610.19331v1 Announce Type: new Abstract: This paper presents LLM-jp-4, a family of Japanese sovereign large language models (LLMs) developed with an emphasis on transparency, reproducibility, and practical usability. LLM-jp-4 comprises an 8B dense model and a 32B-A3B mixtu…