Lego-MT commited on
Commit
6660e70
1 Parent(s): 59f168d

update README

Browse files
Files changed (1) hide show
  1. README.md +12 -12
README.md CHANGED
@@ -1,20 +1,20 @@
1
 
2
  ### Model Sources
3
 
4
- Paper: LlamaX: Scaling Linguistic Horizons of LLM by Enhancing Translation Capabilities Beyond 100 Languages
5
 
6
  Link: https://arxiv.org/pdf/2407
7
 
8
  Repository: https://github.com/CONE-MT/
9
 
10
  ### Model Description
11
- LlamaX is a multilingual language model, developed through continued pre-training on Llama2, and supports over 100 languages.
12
  Its translation capabilities far exceed general models of the same scale, and it can serve as a base model to support downstream multilingual tasks.
13
 
14
 
15
  ### 🔥 Effortless Multilingual Translation with a Simple Prompt
16
 
17
- LlamaX supports translation between more than 100 languages, surpassing the performance of similarly scaled LLMs.
18
 
19
  ```angular2html
20
  def prompt_template(query, src_language, trg_language):
@@ -46,14 +46,14 @@ tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokeniza
46
 
47
  ### 🔥 Effective Base Model for Multilingual Task
48
 
49
- LlamaX preserves its efficacy in general tasks and improves the performance on multilingual tasks.
50
- We fine-tuned LlamaX using only the English training set of downstream task, which also shows significant improvements in non-English. We provide fine-tuning LlamaX models for the following three tasks:
51
 
52
- Math Reasoning: https://huggingface.co/TransLLaMA/TransLLaMA2-7B-MetaMath
53
 
54
- Commonsense Reasoning: https://huggingface.co/TransLLaMA/TransLLaMA2-7B-X-CSQA
55
 
56
- Natural Language Inference: https://huggingface.co/TransLLaMA/TransLLaMA2-7B-XNLI
57
 
58
 
59
  ### Supported Languages
@@ -61,10 +61,10 @@ Akrikaans (af), Amharic (am), Arabic (ar), Armenian (hy), Assamese (as), Asturia
61
 
62
 
63
  ### Model Index
64
- | Model | TransLlama | TransLlama-Alpaca |
65
- |---------|----------------------------------------------------------|-----------------------------------------------------------------|
66
- | Llama-2 | [Link](https://huggingface.co/TransLLaMA/TransLLaMA2-7B) | [Link](https://huggingface.co/TransLLaMA/TransLLaMA2-7B-Alpaca) |
67
- | Llama-3 | [Link](https://huggingface.co/TransLLaMA/TransLLaMA3-8B) | [Link](https://huggingface.co/TransLLaMA/TransLLaMA3-8B-Alpaca) |
68
 
69
  ### Citation
70
  If our model helps your work, please cite this paper:
 
1
 
2
  ### Model Sources
3
 
4
+ Paper: LLaMAX: Scaling Linguistic Horizons of LLM by Enhancing Translation Capabilities Beyond 100 Languages
5
 
6
  Link: https://arxiv.org/pdf/2407
7
 
8
  Repository: https://github.com/CONE-MT/
9
 
10
  ### Model Description
11
+ LLaMAX is a multilingual language model, developed through continued pre-training on Llama2, and supports over 100 languages.
12
  Its translation capabilities far exceed general models of the same scale, and it can serve as a base model to support downstream multilingual tasks.
13
 
14
 
15
  ### 🔥 Effortless Multilingual Translation with a Simple Prompt
16
 
17
+ LLaMAX supports translation between more than 100 languages, surpassing the performance of similarly scaled LLMs.
18
 
19
  ```angular2html
20
  def prompt_template(query, src_language, trg_language):
 
46
 
47
  ### 🔥 Effective Base Model for Multilingual Task
48
 
49
+ LLaMAX preserves its efficacy in general tasks and improves the performance on multilingual tasks.
50
+ We fine-tuned LLaMAX using only the English training set of downstream task, which also shows significant improvements in non-English. We provide fine-tuning LLaMAX models for the following three tasks:
51
 
52
+ Math Reasoning: https://huggingface.co/LLaMAX/LLaMAX2-7B-MetaMath
53
 
54
+ Commonsense Reasoning: https://huggingface.co/LLaMAX/LLaMAX2-7B-X-CSQA
55
 
56
+ Natural Language Inference: https://huggingface.co/LLaMAX/LLaMAX2-7B-XNLI
57
 
58
 
59
  ### Supported Languages
 
61
 
62
 
63
  ### Model Index
64
+ | LLaMAX | LLaMAX-Alpaca |
65
+ |----------------------------------------------------------|-----------------------------------------------------------------|
66
+ | [Link](https://huggingface.co/LLaMAX/LLaMAX2-7B) | [Link](https://huggingface.co/LLaMAX/LLaMAX2-7B-Alpaca) |
67
+ | [Link](https://huggingface.co/LLaMAX/LLaMAX3-8B) | [Link](https://huggingface.co/LLaMAX/LLaMAX3-8B-Alpaca) |
68
 
69
  ### Citation
70
  If our model helps your work, please cite this paper: