kevinpro commited on
Commit
b220808
1 Parent(s): 562362e

commit message

Browse files
Files changed (2) hide show
  1. __pycache__/content.cpython-38.pyc +0 -0
  2. content.py +10 -11
__pycache__/content.cpython-38.pyc CHANGED
Binary files a/__pycache__/content.cpython-38.pyc and b/__pycache__/content.cpython-38.pyc differ
 
content.py CHANGED
@@ -3,25 +3,24 @@ TITLE = '<h1 align="center" id="space-title">Open Multilingual Reasoning Leaderb
3
  INTRO_TEXT = f"""
4
  ## About
5
 
6
- This leaderboard tracks progress and ranks reasoning performance of large language models (LLMs) developed for different languages,
7
- emphasizing on non-English languages to democratize benefits of LLMs to broader society.
8
- Our current leaderboard provides evaluation data for 10 languages.
9
  Both multilingual and language-specific LLMs are welcome in this leaderboard.
10
- We currently evaluate models over four benchmarks:
 
11
 
12
  - <a href="https://huggingface.co/datasets/Mathoctopus/MSVAMP" target="_blank"> MSVAMP </a>
13
  - <a href="https://huggingface.co/datasets/juletxara/mgsm" target="_blank"> MGSM </a>
14
- - <a href="https://arxiv.org/abs/2009.03300" target="_blank"> MNumGLUESub </a>
15
 
16
- # """
17
 
18
- # HOW_TO = f"""
19
- # ## How to list your model performance on this leaderboard:
20
 
21
- # Run the evaluation of your model using this repo: <a href="https://github.com/nlp-uoregon/mlmm-evaluation" target="_blank">https://github.com/nlp-uoregon/mlmm-evaluation</a>.
22
 
23
- # And then, push the evaluation log and make a pull request.
24
- # """
25
 
26
  # CREDIT = f"""
27
  # ## Credit
 
3
  INTRO_TEXT = f"""
4
  ## About
5
 
6
+ This leaderboard tracks and ranks the reasoning performance of the leading, most advanced multilingual reasoning LLMs on three multilingual mathematical reasoning benchmarks. Each benchmark contains 10 languages: Bengali, Swedish, Thailand, Chinese, Japan, Russian, French, Spanish, German and English.
 
 
7
  Both multilingual and language-specific LLMs are welcome in this leaderboard.
8
+
9
+ ## Benchmarks
10
 
11
  - <a href="https://huggingface.co/datasets/Mathoctopus/MSVAMP" target="_blank"> MSVAMP </a>
12
  - <a href="https://huggingface.co/datasets/juletxara/mgsm" target="_blank"> MGSM </a>
13
+ - <a href="https://huggingface.co/datasets/kevinpro/MNumGLUESub" target="_blank"> MNumGLUESub </a>
14
 
15
+ """
16
 
17
+ HOW_TO = f"""
18
+ ## How to list your model performance on this leaderboard:
19
 
20
+ Run the evaluation of your model using this repo: <a href="https://github.com/NJUNLP/MAPO" target="_blank">https://github.com/NJUNLP/MAPO</a>.
21
 
22
+ And then, push the evaluation log and make a pull request.
23
+ """
24
 
25
  # CREDIT = f"""
26
  # ## Credit