add dataset_info in dataset metadata
Browse files
README.md
CHANGED
@@ -20,6 +20,55 @@ paperswithcode_id: null
|
|
20 |
pretty_name: XSum Hallucination Annotations
|
21 |
tags:
|
22 |
- hallucinations
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
23 |
---
|
24 |
|
25 |
# Dataset Card for XSum Hallucination Annotations
|
|
|
20 |
pretty_name: XSum Hallucination Annotations
|
21 |
tags:
|
22 |
- hallucinations
|
23 |
+
dataset_info:
|
24 |
+
- config_name: xsum_factuality
|
25 |
+
features:
|
26 |
+
- name: bbcid
|
27 |
+
dtype: int32
|
28 |
+
- name: system
|
29 |
+
dtype: string
|
30 |
+
- name: summary
|
31 |
+
dtype: string
|
32 |
+
- name: is_factual
|
33 |
+
dtype:
|
34 |
+
class_label:
|
35 |
+
names:
|
36 |
+
0: 'no'
|
37 |
+
1: 'yes'
|
38 |
+
- name: worker_id
|
39 |
+
dtype: string
|
40 |
+
splits:
|
41 |
+
- name: train
|
42 |
+
num_bytes: 800027
|
43 |
+
num_examples: 5597
|
44 |
+
download_size: 2864759
|
45 |
+
dataset_size: 800027
|
46 |
+
- config_name: xsum_faithfulness
|
47 |
+
features:
|
48 |
+
- name: bbcid
|
49 |
+
dtype: int32
|
50 |
+
- name: system
|
51 |
+
dtype: string
|
52 |
+
- name: summary
|
53 |
+
dtype: string
|
54 |
+
- name: hallucination_type
|
55 |
+
dtype:
|
56 |
+
class_label:
|
57 |
+
names:
|
58 |
+
0: intrinsic
|
59 |
+
1: extrinsic
|
60 |
+
- name: hallucinated_span_start
|
61 |
+
dtype: int32
|
62 |
+
- name: hallucinated_span_end
|
63 |
+
dtype: int32
|
64 |
+
- name: worker_id
|
65 |
+
dtype: string
|
66 |
+
splits:
|
67 |
+
- name: train
|
68 |
+
num_bytes: 1750325
|
69 |
+
num_examples: 11185
|
70 |
+
download_size: 2864759
|
71 |
+
dataset_size: 1750325
|
72 |
---
|
73 |
|
74 |
# Dataset Card for XSum Hallucination Annotations
|