README.md

---
license: mit
datasets:
- Umean/B2NERD
language:
- en
- zh
library_name: peft
---

# B2NER 

We present B2NERD, a cohesive and efficient dataset that can improve LLMs' generalization on the challenging Open NER task, refined from 54 existing English or Chinese datasets. 
Our B2NER models, trained on B2NERD, outperform GPT-4 by 6.8-12.0 F1 points and surpass previous methods in 3 out-of-domain benchmarks across 15 datasets and 6 languages.

 - 📖 Paper: [Beyond Boundaries: Learning a Universal Entity Taxonomy across Datasets and Languages for Open Named Entity Recognition](http://arxiv.org/abs/2406.11192)
 - 🎮 Github Repo: https://github.com/UmeanNever/B2NER
 - 📀 Data: See below data section. You can download from [HuggingFace](https://huggingface.co/datasets/Umean/B2NERD) or [Google Drive](https://drive.google.com/file/d/11Wt4RU48i06OruRca2q_MsgpylzNDdjN/view?usp=drive_link).
 - 💾 Model (LoRA Adapters): Current repo saves the B2NER model LoRA adapter based on InternLM2.5-7B. See [20B model](https://huggingface.co/Umean/B2NER-Internlm2-20B-LoRA) for a 20B adapter.

**See github repo for more information about model usage and this work.**

# Cite
```
@article{yang2024beyond,
  title={Beyond Boundaries: Learning a Universal Entity Taxonomy across Datasets and Languages for Open Named Entity Recognition},
  author={Yang, Yuming and Zhao, Wantong and Huang, Caishuang and Ye, Junjie and Wang, Xiao and Zheng, Huiyuan and Nan, Yang and Wang, Yuran and Xu, Xueying and Huang, Kaixin and others},
  journal={arXiv preprint arXiv:2406.11192},
  year={2024}
}
```