Datasets:
Tasks:
Question Answering
Modalities:
Text
Formats:
json
Languages:
Russian
Size:
10K - 100K
Tags:
rag
License:
MLNavigator
commited on
Commit
•
26504ad
1
Parent(s):
34b74cb
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,18 @@
|
|
1 |
---
|
2 |
license: mit
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: mit
|
3 |
---
|
4 |
+
|
5 |
+
Based on Sberquad
|
6 |
+
|
7 |
+
- Answer converted to human affordable answer.
|
8 |
+
- Context augmented with some pices of texts from wiki accordant to text on tematic and keywords.
|
9 |
+
|
10 |
+
- This dataset cold be used for training retrieval LLM models or modificators for ability of LLM to retrieve target information from collection of tematic related texts.
|
11 |
+
|
12 |
+
Dataset consists of 45278 examples of format:
|
13 |
+
{
|
14 |
+
'text': '<text with correct answer>',
|
15 |
+
'q': '<question text>',
|
16 |
+
'a': '<correct answer text>',
|
17 |
+
'context': '<text of 4-10 text chunks, one with right answer and others relevant with text and question on tematic and keywords>'
|
18 |
+
}
|