Datasets:
ZhouChuYue
commited on
Commit
·
7f39997
1
Parent(s):
6a38a85
Merge L1-L3 result note into description paragraph
Browse files- README.md +1 -3
- README_ZH.md +1 -3
README.md
CHANGED
|
@@ -178,14 +178,12 @@ To fairly compare different parsing strategies, we conducted experiments on a da
|
|
| 178 |
|
| 179 |
### Pipeline Effectiveness (L1 vs L2 vs L3)
|
| 180 |
|
| 181 |
-
To validate the effectiveness of our L0-L3 tiered framework, we conducted ablation studies comparing models trained on different tiers of UltraData-Math. Unlike the L0 parser comparison above (which used a 2023-2024 subset), these results are based on the **full dataset**.
|
| 182 |
|
| 183 |
<div align="center">
|
| 184 |
<img src="assets/ultradata-math-l1l2l3-comparison.png" width="700"/>
|
| 185 |
</div>
|
| 186 |
|
| 187 |
-
*Note: Results demonstrate that higher-tier data (L3) significantly boosts mathematical reasoning (MATH500, GSM8K) and general capabilities.*
|
| 188 |
-
|
| 189 |
### Full Evaluation Results
|
| 190 |
|
| 191 |
To compare against existing public mathematical pre-training datasets, we trained models independently on each dataset using the same model architecture and training budget (~100B tokens). The baselines include [Nemotron-CC-Math](https://huggingface.co/datasets/nvidia/Nemotron-CC-Math-v1), [MegaMath-Web-Pro](https://huggingface.co/datasets/LLM360/MegaMath), and [FineMath](https://huggingface.co/datasets/HuggingFaceTB/finemath). All models are evaluated under identical conditions for a fair comparison:
|
|
|
|
| 178 |
|
| 179 |
### Pipeline Effectiveness (L1 vs L2 vs L3)
|
| 180 |
|
| 181 |
+
To validate the effectiveness of our L0-L3 tiered framework, we conducted ablation studies comparing models trained on different tiers of UltraData-Math. Unlike the L0 parser comparison above (which used a 2023-2024 subset), these results are based on the **full dataset**. Results demonstrate that higher-tier data (L3) significantly boosts mathematical reasoning (MATH500, GSM8K) and general capabilities.
|
| 182 |
|
| 183 |
<div align="center">
|
| 184 |
<img src="assets/ultradata-math-l1l2l3-comparison.png" width="700"/>
|
| 185 |
</div>
|
| 186 |
|
|
|
|
|
|
|
| 187 |
### Full Evaluation Results
|
| 188 |
|
| 189 |
To compare against existing public mathematical pre-training datasets, we trained models independently on each dataset using the same model architecture and training budget (~100B tokens). The baselines include [Nemotron-CC-Math](https://huggingface.co/datasets/nvidia/Nemotron-CC-Math-v1), [MegaMath-Web-Pro](https://huggingface.co/datasets/LLM360/MegaMath), and [FineMath](https://huggingface.co/datasets/HuggingFaceTB/finemath). All models are evaluated under identical conditions for a fair comparison:
|
README_ZH.md
CHANGED
|
@@ -139,14 +139,12 @@ ds = load_dataset("openbmb/UltraData-Math", "UltraData-Math-L3-Conversation-Synt
|
|
| 139 |
|
| 140 |
### 流水线有效性(L1 vs L2 vs L3)
|
| 141 |
|
| 142 |
-
为验证 L0-L3 分级框架的有效性,我们对使用不同层级 UltraData-Math 训练的模型进行了消融实验。与上文 L0 解析器对比(使用 2023-2024
|
| 143 |
|
| 144 |
<div align="center">
|
| 145 |
<img src="assets/ultradata-math-l1l2l3-comparison.png" width="700"/>
|
| 146 |
</div>
|
| 147 |
|
| 148 |
-
*注:结果表明,更高层级的数据(L3)显著提升了数学推理能力(MATH500、GSM8K)及通用能力。*
|
| 149 |
-
|
| 150 |
### 完整评测结果
|
| 151 |
|
| 152 |
为与现有公开数学预训练数据集进行对比,我们使用相同的模型架构和训练预算(~100B tokens)在每个数据集上独立训练模型。基线包括 [Nemotron-CC-Math](https://huggingface.co/datasets/nvidia/Nemotron-CC-Math-v1)、[MegaMath-Web-Pro](https://huggingface.co/datasets/LLM360/MegaMath) 和 [FineMath](https://huggingface.co/datasets/HuggingFaceTB/finemath)。所有模型在相同条件下评估以确保公平对比:
|
|
|
|
| 139 |
|
| 140 |
### 流水线有效性(L1 vs L2 vs L3)
|
| 141 |
|
| 142 |
+
为验证 L0-L3 分级框架的有效性,我们对使用不同层级 UltraData-Math 训练的模型进行了消融实验。与上文 L0 解析器对比(使用 2023-2024 子集)不同,以下结果基于**全量数据集**。结果表明,更高层级的数据(L3)显著提升了数学推理能力(MATH500、GSM8K)及通用能力。
|
| 143 |
|
| 144 |
<div align="center">
|
| 145 |
<img src="assets/ultradata-math-l1l2l3-comparison.png" width="700"/>
|
| 146 |
</div>
|
| 147 |
|
|
|
|
|
|
|
| 148 |
### 完整评测结果
|
| 149 |
|
| 150 |
为与现有公开数学预训练数据集进行对比,我们使用相同的模型架构和训练预算(~100B tokens)在每个数据集上独立训练模型。基线包括 [Nemotron-CC-Math](https://huggingface.co/datasets/nvidia/Nemotron-CC-Math-v1)、[MegaMath-Web-Pro](https://huggingface.co/datasets/LLM360/MegaMath) 和 [FineMath](https://huggingface.co/datasets/HuggingFaceTB/finemath)。所有模型在相同条件下评估以确保公平对比:
|