anrilombard/mzansi-text
收藏Hugging Face2026-03-25 更新2026-03-29 收录
下载链接:
https://hf-mirror.com/datasets/anrilombard/mzansi-text
下载链接
链接失效反馈官方服务:
资源简介:
---
language:
- af
- en
- nso
- sot
- ssw
- tsn
- tso
- ven
- xho
- zul
- nbl
tags:
- pretraining
- south-african-languages
- multilingual
- mzansitext
license: apache-2.0
---
# MzansiText
**MzansiText** is a curated multilingual pretraining corpus for all eleven official South African languages.
[](https://github.com/Anri-Lombard/sallm)
[](https://arxiv.org/abs/2603.20732)
[](https://huggingface.co/anrilombard/mzansilm-125m)
[](https://huggingface.co/collections/anrilombard/mzansilm-69635ca7b60efedb9dfcb09e)
## Dataset Details
- Languages: `af`, `en`, `nso`, `sot`, `ssw`, `tsn`, `tso`, `ven`, `xho`, `zul`, `nbl`
- Schema:
```json
{
"text": "string",
"lang": "string"
}
```
- This repository contains the raw train, validation, and test text splits used for the MzansiLM pretraining release.
- The token distribution table below matches the paper-reported corpus statistics.
### Token Distribution (after filtering + 65,536-vocab BPE tokenizer)
| Language | Train Tokens | % | Val Tokens | Test Tokens |
|---|---:|---:|---:|---:|
| Afrikaans | 2,475,913,822 | 64.96 | 1,865,255 | 1,875,605 |
| English | 740,994,679 | 19.44 | 1,813,651 | 1,821,803 |
| isiZulu | 320,224,015 | 8.40 | 2,017,406 | 2,021,343 |
| isiXhosa | 152,212,403 | 3.99 | 2,016,503 | 2,012,000 |
| Sesotho | 97,558,939 | 2.56 | 2,315,298 | 2,316,170 |
| Setswana | 10,082,930 | 0.26 | 1,216,539 | 1,413,473 |
| Sepedi | 6,697,358 | 0.18 | 685,425 | 778,656 |
| Xitsonga | 3,013,408 | 0.08 | 510,463 | 319,496 |
| siSwati | 1,932,989 | 0.05 | 196,247 | 225,810 |
| Tshivenda | 1,852,481 | 0.05 | 191,495 | 243,315 |
| isiNdebele | 818,549 | 0.02 | 106,224 | 143,458 |
| **Total** | **3,811,301,573** | **100** | **12,934,506** | **13,171,129** |
Validation and test sets are capped at approximately 2M tokens per language to prevent high-resource languages from dominating early stopping.
## Usage
```python
from datasets import load_dataset
ds = load_dataset("anrilombard/mzansi-text", split="train")
print(ds[0])
```
## Related Releases
- Paper: [arXiv:2603.20732](https://arxiv.org/abs/2603.20732)
- Model: [anrilombard/mzansilm-125m](https://huggingface.co/anrilombard/mzansilm-125m)
- Tokenized corpus: [anrilombard/mzansi-text-tokenized](https://huggingface.co/datasets/anrilombard/mzansi-text-tokenized)
- GitHub code and configs: [https://github.com/Anri-Lombard/sallm](https://github.com/Anri-Lombard/sallm)
Full preprocessing pipeline (including this exact cleaning script) is in [`data/cleaning/`](https://github.com/Anri-Lombard/sallm/tree/main/data/cleaning) on GitHub.
## Citation
Please cite the paper:
```bibtex
@misc{lombard2026mzansitextmzansilmopencorpus,
title={MzansiText and MzansiLM: An Open Corpus and Decoder-Only Language Model for South African Languages},
author={Anri Lombard and Simbarashe Mawere and Temi Aina and Ethan Wolff and Sbonelo Gumede and Elan Novick and Francois Meyer and Jan Buys},
year={2026},
eprint={2603.20732},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2603.20732},
}
```
## License
Apache License 2.0
language:
- 南非荷兰语(af)
- 英语(en)
- 北索托语(nso)
- 索托语(sot)
- 斯瓦蒂语(ssw)
- 茨瓦纳语(tsn)
- 聪加语(tso)
- 文达语(ven)
- 科萨语(xho)
- 祖鲁语(zul)
- 恩德贝莱语(nbl)
tags:
- 预训练(pretraining)
- 南非语言(south-african-languages)
- 多语言(multilingual)
- mzansitext
license: apache-2.0
# MzansiText
**MzansiText** 是一套经过精心甄选的多语言预训练语料库,涵盖南非全部11种官方语言。
[](https://github.com/Anri-Lombard/sallm)
[](https://arxiv.org/abs/2603.20732)
[](https://huggingface.co/anrilombard/mzansilm-125m)
[](https://huggingface.co/collections/anrilombard/mzansilm-69635ca7b60efedb9dfcb09e)
## 数据集详情
- 支持语言:`af`(南非荷兰语)、`en`(英语)、`nso`(北索托语)、`sot`(索托语)、`ssw`(斯瓦蒂语)、`tsn`(茨瓦纳语)、`tso`(聪加语)、`ven`(文达语)、`xho`(科萨语)、`zul`(祖鲁语)、`nbl`(恩德贝莱语)
- 数据格式(Schema):
json
{
"text": "string",
"lang": "string"
}
本仓库包含用于MzansiLM预训练发布的原始训练、验证与测试文本划分集。下文的分词分布表与论文中报告的语料库统计数据一致。
### 分词分布(经过滤处理 + 65536词表的字节对编码(BPE)分词器)
| 语言 | 训练分词数 | 占比 | 验证分词数 | 测试分词数 |
|---|---:|---:|---:|---:|
| 南非荷兰语(Afrikaans) | 2,475,913,822 | 64.96 | 1,865,255 | 1,875,605 |
| 英语(English) | 740,994,679 | 19.44 | 1,813,651 | 1,821,803 |
| 祖鲁语(isiZulu) | 320,224,015 | 8.40 | 2,017,406 | 2,021,343 |
| 科萨语(isiXhosa) | 152,212,403 | 3.99 | 2,016,503 | 2,012,000 |
| 索托语(Sesotho) | 97,558,939 | 2.56 | 2,315,298 | 2,316,170 |
| 茨瓦纳语(Setswana) | 10,082,930 | 0.26 | 1,216,539 | 1,413,473 |
| 北索托语(Sepedi) | 6,697,358 | 0.18 | 685,425 | 778,656 |
| 聪加语(Xitsonga) | 3,013,408 | 0.08 | 510,463 | 319,496 |
| 斯瓦蒂语(siSwati) | 1,932,989 | 0.05 | 196,247 | 225,810 |
| 文达语(Tshivenda) | 1,852,481 | 0.05 | 191,495 | 243,315 |
| 恩德贝莱语(isiNdebele) | 818,549 | 0.02 | 106,224 | 143,458 |
| **总计** | **3,811,301,573** | **100** | **12,934,506** | **13,171,129** |
验证集与测试集的每语言分词数均限制在约200万,以避免高资源语言主导早停过程。
## 使用方法
python
from datasets import load_dataset
ds = load_dataset("anrilombard/mzansi-text", split="train")
print(ds[0])
## 相关发布
- 论文:[arXiv:2603.20732](https://arxiv.org/abs/2603.20732)
- 模型:[anrilombard/mzansilm-125m](https://huggingface.co/anrilombard/mzansilm-125m)
- 分词后语料库:[anrilombard/mzansi-text-tokenized](https://huggingface.co/datasets/anrilombard/mzansi-text-tokenized)
- GitHub代码与配置:[https://github.com/Anri-Lombard/sallm](https://github.com/Anri-Lombard/sallm)
完整的预处理流水线(包含本数据集所用的清洗脚本)可在GitHub仓库的[`data/cleaning/`](https://github.com/Anri-Lombard/sallm/tree/main/data/cleaning)路径中获取。
## 引用方式
请引用以下论文:
bibtex
@misc{lombard2026mzansitextmzansilmopencorpus,
title={"MzansiText与MzansiLM:面向南非语言的开源语料库与仅解码器语言模型"},
author={Anri Lombard and Simbarashe Mawere and Temi Aina and Ethan Wolff and Sbonelo Gumede and Elan Novick and Francois Meyer and Jan Buys},
year={2026},
eprint={2603.20732},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2603.20732},
}
## 许可证
Apache许可证2.0
提供机构:
anrilombard


