sangmichaelxie commited on
Commit
2420ac9
1 Parent(s): 23d6b75
Files changed (1) hide show
  1. README.md +3 -7
README.md CHANGED
@@ -57,24 +57,20 @@ These chunks define the examples that we do data selection on, totaling 1.7B exa
57
  Before DSIR, we first apply a manual quality filter (see paper for details) and only consider the examples that pass the filter.
58
 
59
 
60
- ### Annotations
61
-
62
- There are no annotations.
63
-
64
-
65
  ## Considerations for Using the Data
66
 
67
  The dataset is biased towards choosing data from non-Wikipedia and non-Books sources. A balanced approach would be to mix in more data from Wikipedia and books.
68
 
69
  ### Dataset Curators
70
 
71
- Sang Michael Xie
72
 
73
  ### Citation Information
 
74
  ```
75
  @article{xie2023data,
76
  author = {Sang Michael Xie and Shibani Santurkar and Tengyu Ma and Percy Liang},
77
- journal = {TODO},
78
  title = {Data Selection for Language Models via Importance Resampling},
79
  year = {2023},
80
  }
 
57
  Before DSIR, we first apply a manual quality filter (see paper for details) and only consider the examples that pass the filter.
58
 
59
 
 
 
 
 
 
60
  ## Considerations for Using the Data
61
 
62
  The dataset is biased towards choosing data from non-Wikipedia and non-Books sources. A balanced approach would be to mix in more data from Wikipedia and books.
63
 
64
  ### Dataset Curators
65
 
66
+ Sang Michael Xie, Shibani Santurkar
67
 
68
  ### Citation Information
69
+ Paper: <https://arxiv.org/abs/2302.03169>
70
  ```
71
  @article{xie2023data,
72
  author = {Sang Michael Xie and Shibani Santurkar and Tengyu Ma and Percy Liang},
73
+ journal = {arXiv preprint arXiv:2302.03169},
74
  title = {Data Selection for Language Models via Importance Resampling},
75
  year = {2023},
76
  }