Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
json
Languages:
English
Size:
10K - 100K
License:
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: mit
|
3 |
configs:
|
|
|
1 |
+
SpanEx consists of 7071 instances annotated for span interactions.
|
2 |
+
SpanEx is the first dataset with human phrase-level interaction explanations with explicit labels for interaction types.
|
3 |
+
Moreover, SpanEx is annotated by three annotators, which opens new avenues for studies of human explanation agreement -- an understudied area in the explainability literature.
|
4 |
+
Our study reveals that while human annotators often agree on span interactions, they also offer complementary reasons for a prediction, collectively providing a comprehensive set of reasons for a prediction.
|
5 |
+
We collect explanations of span interactions for NLI on the SNLI dataset and for FC on the FEVER dataset.
|
6 |
---
|
7 |
license: mit
|
8 |
configs:
|