Fine‑tuned XLM‑RoBERTa‑base on a Japanese NER corpus (tags PER, ORG, LOC, INS, PRD, EVT) using 5‑epoch Adam (lr 5e‑5, batch 12) to reach a 0.0173 validation loss, released on Transformers 4.23.1 and PyTorch 1.12.1.
Description in other languages
Statistics
Statistics as of Nov 16, 2026
Hugging Face downloads (last 30 days)
4,000
Hugging Face likes
28
Listed
—
Listed since the page was created
Downloads (30-day window) over time
Show as a table
| Month | DL (30d) |
|---|---|
| Dec 2025 | 157,000 |
| Jan 2026 | 1,000 |
| Feb 2026 | 2,000 |
| Mar 2026 | 2,000 |
| Apr 2026 | 2,000 |
| May 2026 | 2,000 |
| Jun 2026 | 9,000 |
| Jul 2026 | 6,000 |
| Aug 2026 | 7,000 |
| Sep 2026 | 2,000 |
| Oct 2026 | 3,000 |
| Nov 2026 | 4,000 |
Downloads over the 30 days before the last snapshot of each month.
Popular in the same category
- bert-base-japanese-v3-ner-wikipedia-datasetllm-book8kFine‑tuned Japanese BERT‑Base for named‑entity recognition on a Wikipedia dataset, presented in Chapter 6 of the *Large Language Model Introduction* book and deployable via a Hugging Face transformers pipeline (Apache 2.0 licensed).
- bert-ner-japanesejurabi5kJapanese NER using cl‑tohoku/bert‑base‑japanese‑v2 that extracts eight entity types (corporations, political/other organizations, facilities, products, events) via BertForTokenClassification, trained on the Stockmark Wikipedia dataset and installable with transformers, unidic_lite, and fugashi under a CC BY‑SA 3.0 license.
Resources in similar research fields
bert-japanese-ner-finetuningken1111This is a sample code for creating and using a model for named entity recognition task through finetuning of the BERT model.GitHubShared: Language Models・Semantic Text Processing
novelanalysisbynerlychee12232Named entity extraction using BERT fine-tuningGitHubShared: Information Extraction & Text Mining・Named Entity Recognition
spacy_tutorialyuibi65spaCy tutorial in English and Japanese. spacy-transformers, BERT, GiNZA.GitHubShared: Multilinguality・Language Models- modernbert-ja-310msbintuitions36kModernBERT‑Ja‑310M is a Japanese‑language BERT variant that blends local‑global attention with RoPE, trained on 4.09 T tokens of Japanese/English text, supports a 102 400‑word vocabulary, 8 192‑token sequences, and is optimized for Flash Attention 2.Shared: Language Models・Semantic Text Processing
- deberta-v2-base-japaneseku-nlp24kJapanese DeBERTa V2 base model pretrained on 171 GB of Japanese Wikipedia, CC‑100 and OSCAR data using Juman++ segmentation and SentencePiece tokenization, trained for three weeks on eight NVIDIA A100 GPUs and ready for fine‑tuning.Shared: Language Models・Semantic Text Processing
- modernbert-ja-130msbintuitions12kA 132‑million‑parameter Japanese ModernBERT model that blends local‑global and RoPE attention, trained on 4.39 T tokens (Japanese/English) with a 102‑k‑size vocab, 8,192‑token max length, and optimized for Flash Attention 2.Shared: Language Models・Semantic Text Processing