Text Generation
Transformers
PyTorch
llama
text-generation-inference
Inference Endpoints
File size: 10,670 Bytes
3ac6635
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
---
license: apache-2.0
datasets:
- Mathoctopus/GSM8KInstruct_Parallel
language:
- en
- es
- zh
- de
- ru
- th
- sw
- ja
- fr
- bn
---

# πŸ™ Breaking Language Barriers in Multilingual Mathematical Reasoning: Insights and Observations

Project Page: [https://mathoctopus.github.io/](https://mathoctopus.github.io/)

Paper: [https://arxiv.org/abs/2310.20246.pdf](https://arxiv.org/abs/2310.20246.pdf)

Code: [https://github.com/microsoft/MathOctopus](https://github.com/microsoft/MathOctopus)

### Introduction

We introduce πŸ™ MathOctopus, a series of open-source large language models (LLMs) specifically tailored for multilingual math problem-solving. The MathOctopus models are trained on πŸ€— MGSM8KInstruct Dataset, encompassing ten distinct languages.
MathOctopus notably outperforms conventional open-source LLMs and exhibits superiority over ChatGPT in few-shot scenarios.

### Datasets 

#### **MGSM8KInstruct**

| Training Dataset      | En      | Sw      | Zh      | Bn      | De      | Es      | Fr      | Ja      | Ru      | Th      | Overall |
|:----------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
| MGSM8KInstruct        | 7473    | 7472    | 7466    | 6539    | 7466    | 7470    | 7469    | 7471    | 7361    | 7473    | **73.6K**   |


#### **MSVAMP**

| Test Dataset      | En      | Sw      | Zh      | Bn      | De      | Es      | Fr      | Ja      | Ru      | Th      | Overall |
|:----------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
| MSVAMP                | 1000    | 1000    | 1000    | 1000    | 1000    | 1000    | 1000    | 1000    | 1000    | 1000    | **10K**   |

#### Usage

Our dataset and models are all available at Huggingface.

πŸ€— [MGSM8KInstruct_Parallel Dataset](https://huggingface.co/datasets/Mathoctopus/GSM8KInstruct_Parallel)

πŸ€— [MGSM8KInstruct_Cross Dataset](https://huggingface.co/datasets/Mathoctopus/MGSM8KInstruct_Cross)

πŸ€— [MSVAMP Dataset](https://huggingface.co/datasets/Mathoctopus/MSVAMP)


##  Models

|  Base Model: LLama   	| Parallel-Training                                         	| Cross-Training                                                       	|
|----|---------------------------------------------------------------|---------------------------------------------------------------------------|
| 7B-LLaMA 2  	| πŸ™ [MathOctopus-Parallel-7B](https://huggingface.co/Mathoctopus/Parallel_7B)   	| πŸ™ [MathOctopus-Cross-7B](https://huggingface.co/Mathoctopus/Cross_7B)  	|
|| πŸ™[MathOctopus-Parallel-xRFT-7B](https://huggingface.co/Mathoctopus/Parallel_xRFT_7B)|πŸ™[MathOctopus-Cross-xRFT-7B](https://huggingface.co/Mathoctopus/Cross_xRFT_7B)|
| 13B-LLaMA 2 	| πŸ™ [MathOctopus-Parallel-13B](https://huggingface.co/Mathoctopus/Parallel_13B) 	| πŸ™ [MathOctopus-Cross-13B](https://huggingface.co/Mathoctopus/Cross_13B)	|
|| πŸ™[MathOctopus-Parallel-xRFT-13B](https://huggingface.co/Mathoctopus/Parallel_xRFT_13B)|πŸ™[MathOctopus-Cross-xRFT-13B](https://huggingface.co/Mathoctopus/Cross_xRFT_13B/tree/main)|
| 33B-LLaMA 1 	| πŸ™ [MathOctopus-Parallel-33B](https://huggingface.co/Mathoctopus/Parallel_33B)    | πŸ™ [MathOctopus-Cross-33B](https://huggingface.co/Mathoctopus/Cross_33B/tree/main)  	|
| 70B-LLaMA 2 	| Coming soon!	| Coming Soon!      |

*-Parallel refers to our model trained with the parallel-training strategy. 

*-Cross refers to our model trained with cross-training strategy. 

*-xRFT means we train the model with multilingual rejection sampling.

### **Overall Results on MGSM**

| 7B Model                        | En      | Sw      | Zh      | Bn      | De      | Es      | Fr      | Ja      | Ru      | Th      | Overall |
|:--------------------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
| MathOctopus<sup>C</sup>         | 52.0    | 23.6    | 31.6    | 18.8    | 38.0    | 39.2    | 36.4    | 27.2    | 33.6    | 21.6    | 32.2    |
| **xRFT**-MathOctopus<sup>C</sup>| 51.2    | 24.0    | 33.2    | 18.8    | 36.0    | 41.2    | 37.6    | 29.6    | 36.4    | 25.2    | 33.3    |
| MathOctopus<sup>P</sup>-LoRA    | 30.4    | 15.2    | 23.6    | 10.4    | 22.8    | 24.8    | 26.4    | 18.0    | 22.0    | 14.8    | 20.8    |
| MathOctopus<sup>P</sup>         | 52.4    | 39.2    | 38.4    | 28.8    | 44.8    | 42.4    | 43.6    | 36.0    | 39.6    | 34.4    | 40.0    |
| **xRFT**-MathOctopus<sup>P</sup>| 54.8    | 38.4    | 45.2    | 33.2    | 43.6    | 45.2    | 38.0    | 35.6    | 48.4    | 36.4    | 41.9    |
<p></p >

| 13B Model                       | En      | Sw      | Zh      | Bn      | De      | Es      | Fr      | Ja      | Ru      | Th      | Overall |
|:--------------------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
| MathOctopus<sup>C</sup>         | 56.4    | 27.2    | 39.2    | 24.0    | 47.6    | 49.6    | 47.6    | 40.4    | 42.0    | 24.8    | 39.9    |
| **xRFT**-MathOctopus<sup>C</sup>| 53.6    | 28.0    | 45.2    | 21.2    | 48.0    | 46.4    | 46.0    | 35.2    | 45.6    | 28.8    | 39.8    |
| MathOctopus<sup>P</sup>         | 53.2    | 42.8    | 48.8    | 35.2    | 44.4    | 48.0    | 48.4    | 43.2    | 47.6    | 46.8    | 45.8    |
| **xRFT**-MathOctopus<sup>P</sup>| 51.6    | 46.0    | 51.2    | 42.0    | 49.2    | 53.2    | 49.6    | 39.6    | 47.6    | 46.0    | 47.6    |
<p></p >

| 30-34B Model                    | En      | Sw      | Zh      | Bn      | De      | Es      | Fr      | Ja      | Ru      | Th      | Overall |
|:--------------------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
| MathOctopus<sup>C</sup>         | 55.6    | 24.4    | 36.0    | 19.2    | 40.4    | 51.2    | 44.4    | 27.2    | 37.2    | 21.6    | 35.7    |
| **xRFT**-MathOctopus<sup>C</sup>| 53.6    | 27.6    | 34.4    | 19.2    | 47.2    | 47.6    | 44.8    | 30.8    | 38.8    | 22.8    | 36.7    |
| MathOctopus<sup>P</sup>         | 56.4    | 46.8    | 52.0    | 35.2    | 47.2    | 53.2    | 48.0    | 39.2    | 45.6    | 41.2    | 46.5    |
| **xRFT**-MathOctopus<sup>P</sup>| 51.6    | 47.2    | 52.4    | 37.6    | 51.2    | 52.8    | 44.4    | 41.6    | 50.0    | 47.6    | 47.6    |


### **Overall Results on MSVAMP**

| 7B Model                        | En      | Sw      | Zh      | Bn      | De      | Es      | Fr      | Ja      | Ru      | Th      | Overall |
|:--------------------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
| MathOctopus<sup>C</sup>         | 49.2    | 36.6    | 43.6    | 30.2    | 48.6    | 46.8    | 46.4    | 42.5    | 46.7    | 34.0    | 42.5    |
| **xRFT**-MathOctopus<sup>C</sup>| 49.9    | 37.7    | 43.3    | 32.9    | 46.5    | 47.6    | 47.3    | 42.7    | 46.6    | 36.2    | 43.1    |
| MathOctopus<sup>P</sup>-LoRA    | 30.4    | 15.2    | 23.6    | 10.4    | 22.8    | 24.8    | 26.4    | 18.0    | 22.0    | 14.8    | 20.8    |
| MathOctopus<sup>P</sup>         | 46.5    | 40.1    | 42.5    | 29.1    | 43.5    | 45.4    | 46.0    | 42.5    | 45.4    | 35.7    | 41.7    |
| **xRFT**-MathOctopus<sup>P</sup>| 46.8    | 42.3    | 43.2    | 32.8    | 43.1    | 44.5    | 45.3    | 43.2    | 42.1    | 40.5    | 42.4    |
<p></p >

| 13B Model                       | En      | Sw      | Zh      | Bn      | De      | Es      | Fr      | Ja      | Ru      | Th      | Overall |
|:--------------------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
| MathOctopus<sup>C</sup>         | 56.6    | 40.4    | 49.0    | 30.3    | 50.9    | 54.2    | 54.7    | 46.3    | 52.4    | 35.7    | 47.1    |
| **xRFT**-MathOctopus<sup>C</sup>| 52.9    | 41.9    | 49.2    | 34.1    | 50.5    | 52.8    | 51.5    | 45.8    | 50.2    | 35.7    | 46.5    |
| MathOctopus<sup>P</sup>         | 50.7    | 43.4    | 42.6    | 31.8    | 48.4    | 49.4    | 50.6    | 41.1    | 46.9    | 39.3    | 44.4    |
| **xRFT**-MathOctopus<sup>P</sup>| 44.6    | 43.4    | 46.4    | 34.2    | 47.7    | 48.2    | 49.9    | 43.1    | 48.2    | 39.5    | 44.5    |
<p></p >

| 30-34B Model                    | En      | Sw      | Zh      | Bn      | De      | Es      | Fr      | Ja      | Ru      | Th      | Overall |
|:--------------------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
| MathOctopus<sup>C</sup>         | 51.5    | 42.1    | 46.2    | 23.2    | 50.5    | 52.1    | 52.9    | 42.2    | 50.5    | 33.4    | 44.5    |
| **xRFT**-MathOctopus<sup>C</sup>| 48.1    | 42.8    | 43.6    | 23.3    | 48.7    | 50.0    | 48.9    | 43.4    | 44.6    | 35.5    | 42.9    |
| MathOctopus<sup>P</sup>         | 56.4    | 46.8    | 52.0    | 35.2    | 47.2    | 53.2    | 48.0    | 39.2    | 45.6    | 41.2    | 46.5    |
| **xRFT**-MathOctopus<sup>P</sup>| 48.0    | 42.3    | 46.1    | 36.2    | 47.5    | 48.5    | 48.3    | 45.8    | 47.2    | 41.2    | 45.1    |


### **MathOctopus in English**

| Models                          | GSM8K   | SVAMP   |
|:--------------------------------|:--------|:--------|
| LLaMA 2-7B                      | 42.4    | 38.3    |
| MathOctopus<sup>P</sup>-7B      | 49.3    | 46.8    |
| MathOctopus<sup>C</sup>-7B      | 50.8    | 49.3    |
| LLaMA 2-13B                     | 51.0    | 50.9    |
| MathOctopus<sup>P</sup>-13B     | 55.5    | 52.1    |
| MathOctopus<sup>C</sup>-13B     | 56.6    | 56.6    |
| LLaMA 1-33B                     | 50.0    | 49.0    |
| MathOctopus<sup>P</sup>-33B     | 56.0    | 52.5    |
| MathOctopus<sup>C</sup>-33B     | 53.7    | 51.5    |

## Intended Uses
These models are trained for research purposes. They are designed to solve multilingual math problems. They can be used in educational software, tutoring systems, or any application where a solution to a math problem is needed.

## Citation
Please cite our paper if you use our data, model or code. Please also kindly cite the original dataset papers.

```
@misc{chen2023breaking,
      title={Breaking Language Barriers in Multilingual Mathematical Reasoning: Insights and Observations}, 
      author={Nuo Chen and Zinan Zheng and Ning Wu and Linjun Shou and Ming Gong and Yangqiu Song and Dongmei Zhang and Jia Li},
      year={2023},
      eprint={2310.20246},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}
```