carmentano commited on
Commit
b0b59a9
·
1 Parent(s): 8805d8c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -7
README.md CHANGED
@@ -62,7 +62,9 @@ task_ids:
62
 
63
  CoQCat is a dataset for Conversational Question Answering in Catalan. It is based on [CoQA dataset](https://stanfordnlp.github.io/coqa/).
64
 
65
- CoQCat comprises 89384 question-answer pairs, sourced from conversations related to text passages from six different domains. The questions and responses are designed to maintain a conversational tone. The answers are presented in a free-form text format, with evidence highlighted from the passage.
 
 
66
 
67
  For the development and test sets, an additional 2 responses to each question have been collected.
68
 
@@ -203,12 +205,12 @@ We created this dataset to contribute to the development of language models in C
203
 
204
  We obtained the initial text passages from different sources, depending on the domain:
205
 
206
- * biographies: [Catalan wikipedia](ca.wikipedia.org)
207
- * litterature: [Gutenberg Project](gutenberg.org/)
208
- * news: [VilaWeb](vilaweb.cat)
209
- * mitology: [Catalan wikipedia](ca.wikipedia.org)
210
- * short histories (only in `dev`and `test`): [Petites històries](https://petiteshistories.wordpress.com/)
211
- * movie plot (only in `dev`and `test`): [Catalan wikipedia](ca.wikipedia.org)
212
 
213
  #### Who are the source language producers?
214
 
 
62
 
63
  CoQCat is a dataset for Conversational Question Answering in Catalan. It is based on [CoQA dataset](https://stanfordnlp.github.io/coqa/).
64
 
65
+ CoQCat comprises 89384 question-answer pairs, sourced from conversations related to text 6000 passages from six different domains.
66
+ The questions and responses are designed to maintain a conversational tone.
67
+ The answers are presented in a free-form text format, with evidence highlighted from the passage.
68
 
69
  For the development and test sets, an additional 2 responses to each question have been collected.
70
 
 
205
 
206
  We obtained the initial text passages from different sources, depending on the domain:
207
 
208
+ * biographies: [Catalan wikipedia](ca.wikipedia.org) - 1450 text passages
209
+ * litterature: [Gutenberg Project](gutenberg.org/) - 1450 text passages
210
+ * news: [VilaWeb](vilaweb.cat) - 1450 text passages
211
+ * mitology: [Catalan wikipedia](ca.wikipedia.org) - 1450 text passages
212
+ * short histories (only in `test`): [Petites històries](https://petiteshistories.wordpress.com/) - 100 text passages
213
+ * movie plot (only in `test`): [Catalan wikipedia](ca.wikipedia.org) - 100 text passages
214
 
215
  #### Who are the source language producers?
216