Datasets:
Update README.md
#2
by
spuriousmllm
- opened
README.md
CHANGED
|
@@ -2,6 +2,10 @@
|
|
| 2 |
license: mit
|
| 3 |
task_categories:
|
| 4 |
- question-answering
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
---
|
| 6 |
# MM-SpuBench Datacard
|
| 7 |
|
|
@@ -27,5 +31,5 @@ task_categories:
|
|
| 27 |
|
| 28 |
- **Content Types:** VQA pairs with Concept Information
|
| 29 |
- **Volume:** 2400 VQA pairs
|
| 30 |
-
- **Source of Data:**
|
| 31 |
- **Data Collection Method:** See the paper.
|
|
|
|
| 2 |
license: mit
|
| 3 |
task_categories:
|
| 4 |
- question-answering
|
| 5 |
+
language:
|
| 6 |
+
- en
|
| 7 |
+
size_categories:
|
| 8 |
+
- 1K<n<10K
|
| 9 |
---
|
| 10 |
# MM-SpuBench Datacard
|
| 11 |
|
|
|
|
| 31 |
|
| 32 |
- **Content Types:** VQA pairs with Concept Information
|
| 33 |
- **Volume:** 2400 VQA pairs
|
| 34 |
+
- **Source of Data:** Images derived from ObjectNet as the primary source, and ImageNet-R (rendition), ImageNet-Sketch, ImageNet-A, and ImageNet-C (using ImageNet-Hard subset). QA data are curated by the authors.
|
| 35 |
- **Data Collection Method:** See the paper.
|