Datasets:

Modalities:
Text
Formats:
parquet
ArXiv:
Libraries:
Datasets
Dask
License:
Research-EAI commited on
Commit
87df0bb
·
verified ·
1 Parent(s): e07a3f9

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +0 -4
README.md CHANGED
@@ -41,10 +41,6 @@ Our approach simplifies math dataset creation:
41
  1. **Traditional Method**: Train high-recall classifiers → Run on billions of documents
42
  2. **Our Method**: Query taxonomy metadata for `51 - Mathematics` → Apply FineMath classifier to all recalled documents → Select top-scoring content
43
 
44
- ### Taxonomy Category Used:
45
- - **Primary**: `51 - Mathematics` (FDC taxonomy label)
46
- - **Quality Filtering**: FineMath Classifier applied to all 116M recalled documents, selecting top 34B tokens
47
-
48
  ## 🎓 Citation
49
 
50
  If you use this dataset, please cite our EssentialWeb paper:
 
41
  1. **Traditional Method**: Train high-recall classifiers → Run on billions of documents
42
  2. **Our Method**: Query taxonomy metadata for `51 - Mathematics` → Apply FineMath classifier to all recalled documents → Select top-scoring content
43
 
 
 
 
 
44
  ## 🎓 Citation
45
 
46
  If you use this dataset, please cite our EssentialWeb paper: