pszemraj commited on
Commit
11e3d4d
1 Parent(s): 031526e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +23 -0
README.md CHANGED
@@ -6,6 +6,29 @@ source_datasets: tau/scrolls
6
  ---
7
  # qmsum-cleaned
8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9
  ## token counts
10
 
11
  ![counts](https://i.imgur.com/rARAOvr.png)
 
6
  ---
7
  # qmsum-cleaned
8
 
9
+
10
+
11
+ ## prefixes
12
+
13
+
14
+ It's worth noting that each "document" in `input` is prefixed by a question/prompt on what the model is supposed to do. **You may want to explicitly handle this in some way, or prefix your models trained on this dataset.**
15
+
16
+
17
+ Most frequent "prefixes" separated via [sentence-splitter](https://github.com/mediacloud/sentence-splitter) in the `train` split:
18
+
19
+ | | Sentence | Count |
20
+ |---:|:------------------------------------------------------------------------------|--------:|
21
+ | 0 | Summarize the whole meeting. | 121 |
22
+ | 1 | Summarize the meeting | 25 |
23
+ | 2 | What did the team discuss about the product cost? | 4 |
24
+ | 3 | How did Marketing design the product evaluation? | 4 |
25
+ | 4 | Summarize the wrap up of the meeting. | 3 |
26
+ | 5 | What did the group discuss about user requirements of the new remote control? | 3 |
27
+ | 6 | What did the team discuss during the product evaluation? | 3 |
28
+ | 7 | Summarize the meeting. | 2 |
29
+ | 8 | Summarize what was said about digits form | 2 |
30
+ | 9 | What was discussed in the meeting? | 2 |
31
+
32
  ## token counts
33
 
34
  ![counts](https://i.imgur.com/rARAOvr.png)