Commit
β’
bec41fd
1
Parent(s):
4c4c10f
Update README.md
Browse files
README.md
CHANGED
@@ -14,6 +14,7 @@ language:
|
|
14 |
- fr
|
15 |
- bn
|
16 |
---
|
|
|
17 |
### Introduction
|
18 |
|
19 |
We introduce π MathOctopus, a series of open-source large language models (LLMs) specifically tailored for multilingual math problem-solving. The MathOctopus models are trained on π€ MGSM8KInstruct Dataset, encompassing ten distinct languages.
|
@@ -49,9 +50,9 @@ Our dataset and models are all available at Huggingface.
|
|
49 |
|----|---------------------------------------------------------------|---------------------------------------------------------------------------|
|
50 |
| 7B-LLaMA 2 | π [MathOctopus-Parallel-7B](https://huggingface.co/Mathoctopus/Parallel_7B) | π [MathOctopus-Cross-7B](https://huggingface.co/Mathoctopus/Cross_7B) |
|
51 |
|| π[MathOctopus-Parallel-xRFT-7B](https://huggingface.co/Mathoctopus/Parallel_xRFT_7B)|π[MathOctopus-Cross-xRFT-7B](https://huggingface.co/Mathoctopus/Cross_xRFT_7B)|
|
52 |
-
| 13B-LLaMA 2 | π [MathOctopus-Parallel-13B] | π [MathOctopus-Cross-13B] |
|
53 |
-
|| π[MathOctopus-Parallel-xRFT-13B](https://huggingface.co/Mathoctopus/Parallel_xRFT_13B
|
54 |
-
| 33B-LLaMA 1 | π [MathOctopus-Parallel-33B] | π [MathOctopus-Cross-33B] |
|
55 |
| 70B-LLaMA 2 | Coming soon! | Coming Soon! |
|
56 |
|
57 |
*-Parallel refers to our model trained with the parallel-training strategy.
|
@@ -59,7 +60,9 @@ Our dataset and models are all available at Huggingface.
|
|
59 |
*-Cross refers to our model trained with cross-training strategy.
|
60 |
|
61 |
*-xRFT means we train the model with multilingual rejection sampling.
|
|
|
62 |
### **Overall Results on MGSM**
|
|
|
63 |
| 7B Model | En | Sw | Zh | Bn | De | Es | Fr | Ja | Ru | Th | Overall |
|
64 |
|:--------------------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
|
65 |
| 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 |
|
@@ -83,7 +86,10 @@ Our dataset and models are all available at Huggingface.
|
|
83 |
| **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 |
|
84 |
| 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 |
|
85 |
| **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 |
|
|
|
|
|
86 |
### **Overall Results on MSVAMP**
|
|
|
87 |
| 7B Model | En | Sw | Zh | Bn | De | Es | Fr | Ja | Ru | Th | Overall |
|
88 |
|:--------------------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
|
89 |
| 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 |
|
@@ -107,7 +113,10 @@ Our dataset and models are all available at Huggingface.
|
|
107 |
| **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 |
|
108 |
| 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 |
|
109 |
| **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 |
|
|
|
|
|
110 |
### **MathOctopus in English**
|
|
|
111 |
| Models | GSM8K | SVAMP |
|
112 |
|:--------------------------------|:--------|:--------|
|
113 |
| LLaMA 2-7B | 42.4 | 38.3 |
|
@@ -119,5 +128,6 @@ Our dataset and models are all available at Huggingface.
|
|
119 |
| LLaMA 1-33B | 50.0 | 49.0 |
|
120 |
| MathOctopus<sup>P</sup>-33B | 56.0 | 52.5 |
|
121 |
| MathOctopus<sup>C</sup>-33B | 53.7 | 51.5 |
|
|
|
122 |
## Intended Uses
|
123 |
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.
|
|
|
14 |
- fr
|
15 |
- bn
|
16 |
---
|
17 |
+
|
18 |
### Introduction
|
19 |
|
20 |
We introduce π MathOctopus, a series of open-source large language models (LLMs) specifically tailored for multilingual math problem-solving. The MathOctopus models are trained on π€ MGSM8KInstruct Dataset, encompassing ten distinct languages.
|
|
|
50 |
|----|---------------------------------------------------------------|---------------------------------------------------------------------------|
|
51 |
| 7B-LLaMA 2 | π [MathOctopus-Parallel-7B](https://huggingface.co/Mathoctopus/Parallel_7B) | π [MathOctopus-Cross-7B](https://huggingface.co/Mathoctopus/Cross_7B) |
|
52 |
|| π[MathOctopus-Parallel-xRFT-7B](https://huggingface.co/Mathoctopus/Parallel_xRFT_7B)|π[MathOctopus-Cross-xRFT-7B](https://huggingface.co/Mathoctopus/Cross_xRFT_7B)|
|
53 |
+
| 13B-LLaMA 2 | π [MathOctopus-Parallel-13B](https://huggingface.co/Mathoctopus/Parallel_13B) | π [MathOctopus-Cross-13B](https://huggingface.co/Mathoctopus/Cross_13B) |
|
54 |
+
|| π[MathOctopus-Parallel-xRFT-13B](https://huggingface.co/Mathoctopus/Parallel_xRFT_13B)|π[MathOctopus-Cross-xRFT-13B]|
|
55 |
+
| 33B-LLaMA 1 | π [MathOctopus-Parallel-33B](https://huggingface.co/Mathoctopus/Parallel_33B) | π [MathOctopus-Cross-33B] |
|
56 |
| 70B-LLaMA 2 | Coming soon! | Coming Soon! |
|
57 |
|
58 |
*-Parallel refers to our model trained with the parallel-training strategy.
|
|
|
60 |
*-Cross refers to our model trained with cross-training strategy.
|
61 |
|
62 |
*-xRFT means we train the model with multilingual rejection sampling.
|
63 |
+
|
64 |
### **Overall Results on MGSM**
|
65 |
+
|
66 |
| 7B Model | En | Sw | Zh | Bn | De | Es | Fr | Ja | Ru | Th | Overall |
|
67 |
|:--------------------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
|
68 |
| 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 |
|
|
|
86 |
| **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 |
|
87 |
| 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 |
|
88 |
| **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 |
|
89 |
+
|
90 |
+
|
91 |
### **Overall Results on MSVAMP**
|
92 |
+
|
93 |
| 7B Model | En | Sw | Zh | Bn | De | Es | Fr | Ja | Ru | Th | Overall |
|
94 |
|:--------------------------------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|:--------|
|
95 |
| 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 |
|
|
|
113 |
| **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 |
|
114 |
| 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 |
|
115 |
| **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 |
|
116 |
+
|
117 |
+
|
118 |
### **MathOctopus in English**
|
119 |
+
|
120 |
| Models | GSM8K | SVAMP |
|
121 |
|:--------------------------------|:--------|:--------|
|
122 |
| LLaMA 2-7B | 42.4 | 38.3 |
|
|
|
128 |
| LLaMA 1-33B | 50.0 | 49.0 |
|
129 |
| MathOctopus<sup>P</sup>-33B | 56.0 | 52.5 |
|
130 |
| MathOctopus<sup>C</sup>-33B | 53.7 | 51.5 |
|
131 |
+
|
132 |
## Intended Uses
|
133 |
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.
|