Update README.md
Browse files
README.md
CHANGED
@@ -1,14 +1,26 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
4 |
## MAPO: Advancing Multilingual Reasoning through Multilingual Alignment-as-Preference Optimization
|
5 |
-
see our paper in https://arxiv.org/abs/2401.06838
|
6 |
|
|
|
|
|
|
|
|
|
|
|
7 |
|
8 |
-
View the Github Project:
|
9 |
https://github.com/NJUNLP/MAPO
|
10 |
|
11 |
-
Multilingual Reasoning
|
|
|
|
|
|
|
|
|
12 |
|
13 |
## Benchmarks
|
14 |
| System | [MSVAMP](https://huggingface.co/datasets/Mathoctopus/MSVAMP) | [MGSM](https://huggingface.co/datasets/juletxara/mgsm) | [MNumGLUESub](https://huggingface.co/datasets/Mathoctopus/MSVAMP) |
|
@@ -22,8 +34,8 @@ Multilingual Reasoning Benchmark: https://huggingface.co/spaces/kevinpro/Open-Mu
|
|
22 |
| **[+ MAPO-DPO(ours)🔥](https://huggingface.co/kevinpro/MathOctopus-MAPO-DPO-7B)** | **57.4** | **41.6** | **50.4** |
|
23 |
| [MetaMathOctopus 7B](https://huggingface.co/kevinpro/MetaMathOctopus-7B) | 53.0 | 45.5 | 39.2 |
|
24 |
| **[+ MAPO-DPO(ours) 👑](https://huggingface.co/kevinpro/MetaMathOctopus-MAPO-DPO-7B)** | **64.7** | **51.6** | **52.9** |
|
25 |
-
| MistralMathOctopus 7B | 59.0 | 58.0 | 56.8 |
|
26 |
-
|
|
27 |
|
28 |
| System | [MSVAMP](https://huggingface.co/datasets/Mathoctopus/MSVAMP) | [MGSM](https://huggingface.co/datasets/juletxara/mgsm) | [MNumGLUESub](https://huggingface.co/datasets/Mathoctopus/MSVAMP) |
|
29 |
| ------ | :-------------------------------------------------------------: | :------------------------------------------------------: | :-------------------------------------------------------------: |
|
@@ -50,4 +62,4 @@ If you find this model helpful, feel free to cite our paper:
|
|
50 |
archivePrefix={arXiv},
|
51 |
primaryClass={cs.CL}
|
52 |
}
|
53 |
-
```
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
+
metrics:
|
4 |
+
- accuracy
|
5 |
+
pipeline_tag: text-generation
|
6 |
+
tags:
|
7 |
+
- multilingual
|
8 |
---
|
9 |
## MAPO: Advancing Multilingual Reasoning through Multilingual Alignment-as-Preference Optimization
|
|
|
10 |
|
11 |
+
**🔥Our paper**
|
12 |
+
|
13 |
+
https://arxiv.org/abs/2401.06838
|
14 |
+
|
15 |
+
**🔥Github Project**
|
16 |
|
|
|
17 |
https://github.com/NJUNLP/MAPO
|
18 |
|
19 |
+
**🔥Open Multilingual Reasoning Leaderboard**
|
20 |
+
|
21 |
+
https://huggingface.co/spaces/kevinpro/Open-Multilingual-Reasoning-Leaderboard
|
22 |
+
|
23 |
+
|
24 |
|
25 |
## Benchmarks
|
26 |
| System | [MSVAMP](https://huggingface.co/datasets/Mathoctopus/MSVAMP) | [MGSM](https://huggingface.co/datasets/juletxara/mgsm) | [MNumGLUESub](https://huggingface.co/datasets/Mathoctopus/MSVAMP) |
|
|
|
34 |
| **[+ MAPO-DPO(ours)🔥](https://huggingface.co/kevinpro/MathOctopus-MAPO-DPO-7B)** | **57.4** | **41.6** | **50.4** |
|
35 |
| [MetaMathOctopus 7B](https://huggingface.co/kevinpro/MetaMathOctopus-7B) | 53.0 | 45.5 | 39.2 |
|
36 |
| **[+ MAPO-DPO(ours) 👑](https://huggingface.co/kevinpro/MetaMathOctopus-MAPO-DPO-7B)** | **64.7** | **51.6** | **52.9** |
|
37 |
+
| [MistralMathOctopus 7B](https://huggingface.co/kevinpro/MistralMathOctopus-7B) | 59.0 | 58.0 | 56.8 |
|
38 |
+
| **[+ MAPO-DPO(ours) 👑](https://huggingface.co/kevinpro/MistralMathOctopus-MAPO-DPO-7B)** | **74.6** | **67.3** | **70.0** |
|
39 |
|
40 |
| System | [MSVAMP](https://huggingface.co/datasets/Mathoctopus/MSVAMP) | [MGSM](https://huggingface.co/datasets/juletxara/mgsm) | [MNumGLUESub](https://huggingface.co/datasets/Mathoctopus/MSVAMP) |
|
41 |
| ------ | :-------------------------------------------------------------: | :------------------------------------------------------: | :-------------------------------------------------------------: |
|
|
|
62 |
archivePrefix={arXiv},
|
63 |
primaryClass={cs.CL}
|
64 |
}
|
65 |
+
```
|