add German and Arabic
Browse files
README.md
CHANGED
@@ -7,6 +7,8 @@ language:
|
|
7 |
- fr
|
8 |
- ru
|
9 |
- es
|
|
|
|
|
10 |
tags:
|
11 |
- medical
|
12 |
size_categories:
|
@@ -16,6 +18,8 @@ size_categories:
|
|
16 |
[💻Github Repo](https://github.com/MAGIC-AI4Med/MMedLM) [🖨️arXiv Paper](https://arxiv.org/abs/2402.13963)
|
17 |
|
18 |
The official pre-training dataset for "Towards Building Multilingual Language Model for Medicine".
|
|
|
|
|
19 |
|
20 |
## Introduction
|
21 |
This repo contains MMedC, a multilingual medical corpus with 25.5 billion tokens.
|
@@ -27,6 +31,8 @@ This repo contains MMedC, a multilingual medical corpus with 25.5 billion tokens
|
|
27 |
| Russian | Indo-European | 1.29 | 0.40 | 0.00 | 0.00 | 1.69 |
|
28 |
| Chinese | Sino-Tibetan | 3.34 | 1.21 | 0.00 | 0.19 | 4.74 |
|
29 |
| Japaneses | Sino-Tibetan | 1.93 | 0.00 | 0.10 | 0.01 | 2.05 |
|
|
|
|
|
30 |
|
31 |
- English Textbooks is not included in this repo due to copyright issues. For this part of 4B English corpus, please refer to [PMC-LLaMA](https://github.com/chaoyi-wu/PMC-LLaMA)
|
32 |
|
|
|
7 |
- fr
|
8 |
- ru
|
9 |
- es
|
10 |
+
- ar
|
11 |
+
- de
|
12 |
tags:
|
13 |
- medical
|
14 |
size_categories:
|
|
|
18 |
[💻Github Repo](https://github.com/MAGIC-AI4Med/MMedLM) [🖨️arXiv Paper](https://arxiv.org/abs/2402.13963)
|
19 |
|
20 |
The official pre-training dataset for "Towards Building Multilingual Language Model for Medicine".
|
21 |
+
## News
|
22 |
+
- We add Arabic and German corpus to MMedC.
|
23 |
|
24 |
## Introduction
|
25 |
This repo contains MMedC, a multilingual medical corpus with 25.5 billion tokens.
|
|
|
31 |
| Russian | Indo-European | 1.29 | 0.40 | 0.00 | 0.00 | 1.69 |
|
32 |
| Chinese | Sino-Tibetan | 3.34 | 1.21 | 0.00 | 0.19 | 4.74 |
|
33 |
| Japaneses | Sino-Tibetan | 1.93 | 0.00 | 0.10 | 0.01 | 2.05 |
|
34 |
+
| Arabic | Afro-Asiatic | 0.64 | 0.00 | 0.00 | 0.00 | 0.64 |
|
35 |
+
| German | Indo-European | 1.54 | 0.00 | 0.00 | 0.00 | 1.54 |
|
36 |
|
37 |
- English Textbooks is not included in this repo due to copyright issues. For this part of 4B English corpus, please refer to [PMC-LLaMA](https://github.com/chaoyi-wu/PMC-LLaMA)
|
38 |
|