Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,7 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
This dataset contains the following fields:
|
| 2 |
|
| 3 |
- **id**: The unique identifier of the sample.
|
| 4 |
- **question**: The specific prompt, containing the control rules for each text and image.
|
| 5 |
- **content**: The image path or the text content that needs to be reasoned over.
|
| 6 |
- **options**: The human-annotated ground-truth.
|
| 7 |
-
- **extra**: Additional information.
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
task_categories:
|
| 4 |
+
- question-answering
|
| 5 |
+
- text-classification
|
| 6 |
+
language:
|
| 7 |
+
- zh
|
| 8 |
+
tags:
|
| 9 |
+
- E-commerce
|
| 10 |
+
- Benchmark
|
| 11 |
+
- Adversarial Samples
|
| 12 |
+
- Large Language Models
|
| 13 |
+
size_categories:
|
| 14 |
+
- 10K<n<100K
|
| 15 |
+
---
|
| 16 |
This dataset contains the following fields:
|
| 17 |
|
| 18 |
- **id**: The unique identifier of the sample.
|
| 19 |
- **question**: The specific prompt, containing the control rules for each text and image.
|
| 20 |
- **content**: The image path or the text content that needs to be reasoned over.
|
| 21 |
- **options**: The human-annotated ground-truth.
|
| 22 |
+
- **extra**: Additional information.
|