Commit
•
a511458
1
Parent(s):
e120fcc
Update README.md
Browse files
README.md
CHANGED
|
@@ -7,8 +7,6 @@ language:
|
|
| 7 |
- en
|
| 8 |
---
|
| 9 |
<h1 align="center">
|
| 10 |
-
<img src="./assets/jiuzhang.png" width="100" alt="JiuZhang" />
|
| 11 |
-
<br>
|
| 12 |
JiuZhang3.0: Efficiently Improving Mathematical
|
| 13 |
Reasoning by Training Small Data Synthesis Models
|
| 14 |
</h1>
|
|
@@ -22,10 +20,7 @@ Reasoning by Training Small Data Synthesis Models
|
|
| 22 |
## Introduction
|
| 23 |
JiuZhang3.0 is a series of fine-tuned models for math reasoning continually pre-trained on corpus synthesized by our carefully trained small LLM.
|
| 24 |
|
| 25 |
-
|
| 26 |
-
<img src="./assets/model.png" alt="JiuZhang3.0" />
|
| 27 |
-
|
| 28 |
-
### Experimental Results
|
| 29 |
For more evaluation results, please refer to the [Paper](https://arxiv.org/abs/2405.14365)
|
| 30 |
|
| 31 |
| Models | GSM8k | MATH | SVAMP | ASDiv | MAWPS | CARP | Avg. |
|
|
|
|
| 7 |
- en
|
| 8 |
---
|
| 9 |
<h1 align="center">
|
|
|
|
|
|
|
| 10 |
JiuZhang3.0: Efficiently Improving Mathematical
|
| 11 |
Reasoning by Training Small Data Synthesis Models
|
| 12 |
</h1>
|
|
|
|
| 20 |
## Introduction
|
| 21 |
JiuZhang3.0 is a series of fine-tuned models for math reasoning continually pre-trained on corpus synthesized by our carefully trained small LLM.
|
| 22 |
|
| 23 |
+
## Experimental Results
|
|
|
|
|
|
|
|
|
|
| 24 |
For more evaluation results, please refer to the [Paper](https://arxiv.org/abs/2405.14365)
|
| 25 |
|
| 26 |
| Models | GSM8k | MATH | SVAMP | ASDiv | MAWPS | CARP | Avg. |
|