Datasets:
ZhouChuYue
commited on
Commit
·
f73fc34
1
Parent(s):
b9a2c27
Update README: fix L0 experiment description
Browse files- README.md +1 -1
- README_ZH.md +1 -1
README.md
CHANGED
|
@@ -162,7 +162,7 @@ We evaluated data quality using the **Decay Verification** method: continuing pr
|
|
| 162 |
|
| 163 |
### Effectiveness of L0 Parsing Strategy
|
| 164 |
|
| 165 |
-
To fairly compare different parsing strategies, we conducted experiments on a data subset sampled from the **2023-2024** distribution. We re-parsed the raw HTML from this source using different parsers
|
| 166 |
|
| 167 |
| Parser | Average | MMLU | MMLU-STEM | MATH500 | GSM8K | MBPP | HumanEval |
|
| 168 |
|:---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|
|
|
|
| 162 |
|
| 163 |
### Effectiveness of L0 Parsing Strategy
|
| 164 |
|
| 165 |
+
To fairly compare different parsing strategies, we conducted experiments on a data subset sampled from the **2023-2024** distribution. We re-parsed the raw HTML from this source using different parsers. This comparison demonstrates the **effectiveness of our L0 Parser** against other parsers.
|
| 166 |
|
| 167 |
| Parser | Average | MMLU | MMLU-STEM | MATH500 | GSM8K | MBPP | HumanEval |
|
| 168 |
|:---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|
README_ZH.md
CHANGED
|
@@ -127,7 +127,7 @@ ds = load_dataset("openbmb/UltraData-Math", "UltraData-Math-L3-Conversation-Synt
|
|
| 127 |
|
| 128 |
### L0 解析策略有效性
|
| 129 |
|
| 130 |
-
为公平对比不同解析策略,我们在 **2023-2024** 年分布的数据子集上进行实验。我们使用不同解析器重新解析原始 HTML
|
| 131 |
|
| 132 |
| 解析器 | 平均分 | MMLU | MMLU-STEM | MATH500 | GSM8K | MBPP | HumanEval |
|
| 133 |
|:---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|
|
|
|
| 127 |
|
| 128 |
### L0 解析策略有效性
|
| 129 |
|
| 130 |
+
为公平对比不同解析策略,我们在 **2023-2024** 年分布的数据子集上进行实验。我们使用不同解析器重新解析原始 HTML。该对比展示了我们 **L0 解析器的有效性**。
|
| 131 |
|
| 132 |
| 解析器 | 平均分 | MMLU | MMLU-STEM | MATH500 | GSM8K | MBPP | HumanEval |
|
| 133 |
|:---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|