Datasets:
GEM
/

Tasks:
Other
Modalities:
Text
Languages:
English
ArXiv:
Libraries:
Datasets
License:
chandrab commited on
Commit
3e8d3c5
1 Parent(s): 0c60eaa

Create dataset_card.json

Browse files
Files changed (1) hide show
  1. dataset_card.json +163 -0
dataset_card.json ADDED
@@ -0,0 +1,163 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "overview": {
3
+ "where": {
4
+ "has-leaderboard": "no",
5
+ "leaderboard-url": "N/A",
6
+ "leaderboard-description": "N/A",
7
+ "website": "http://abductivecommonsense.xyz/",
8
+ "data-url": "https://storage.googleapis.com/ai2-mosaic/public/abductive-commonsense-reasoning-iclr2020/anlg.zip",
9
+ "paper-url": "https://openreview.net/pdf?id=Byg1v1HKDB",
10
+ "paper-bibtext": "@inproceedings{\nBhagavatula2020Abductive,\ntitle={Abductive Commonsense Reasoning},\nauthor={Chandra Bhagavatula and Ronan Le Bras and Chaitanya Malaviya and Keisuke Sakaguchi and Ari Holtzman and Hannah Rashkin and Doug Downey and Wen-tau Yih and Yejin Choi},\nbooktitle={International Conference on Learning Representations},\nyear={2020},\nurl={https://openreview.net/forum?id=Byg1v1HKDB}\n}",
11
+ "contact-name": "Chandra Bhagavatulla",
12
+ "contact-email": "chandrab@allenai.org"
13
+ },
14
+ "languages": {
15
+ "is-multilingual": "no",
16
+ "license": "apache-2.0: Apache License 2.0",
17
+ "task-other": "N/A",
18
+ "language-names": [
19
+ "English"
20
+ ],
21
+ "language-speakers": "Crowdworkers on the Amazon Mechanical Turk platform based in the U.S, Canada, U.K and Australia. ",
22
+ "intended-use": "To study the viability of language-based abductive reasoning. Training and evaluating models to generate a plausible hypothesis to explain two given observations.",
23
+ "license-other": "N/A",
24
+ "task": "Reasoning"
25
+ },
26
+ "credit": {
27
+ "organization-type": [
28
+ "industry"
29
+ ],
30
+ "organization-names": "Allen Institute for AI",
31
+ "creators": "Chandra Bhagavatula (AI2), Ronan Le Bras (AI2), Chaitanya Malaviya (AI2), Keisuke Sakaguchi (AI2), Ari Holtzman (AI2, UW), Hannah Rashkin (AI2, UW), Doug Downey (AI2), Wen-tau Yih (AI2), Yejin Choi (AI2, UW)",
32
+ "funding": "Allen Institute for AI",
33
+ "gem-added-by": "Chandra Bhagavatula (AI2), Ronan LeBras (AI2), Aman Madaan (CMU), Nico Daheim (RWTH Aachen University)"
34
+ },
35
+ "structure": {
36
+ "data-fields": "- observation_1: A string describing an observation / event.\n- observation_2: A string describing an observation / event.\n- label: A string that plausibly explains why observation_1 and observation_2 might have happened.",
37
+ "structure-labels": "Explanations were authored by crowdworkers on the Amazon Mechanical Turk platform using a custom template designed by the creators of the dataset.",
38
+ "structure-example": "{\n'gem_id': 'GEM-ART-validation-0',\n'observation_1': 'Stephen was at a party.',\n'observation_2': 'He checked it but it was completely broken.',\n'label': 'Stephen knocked over a vase while drunk.'\n}",
39
+ "structure-splits": "- train: Consists of training instances. \n- dev: Consists of dev instances.\n- test: Consists of test instances.\n"
40
+ }
41
+ },
42
+ "gem": {
43
+ "rationale": {
44
+ "contribution": "Abductive reasoning is a crucial capability of humans and ART is the first dataset curated to study language-based abductive reasoning.",
45
+ "sole-task-dataset": "no",
46
+ "distinction-description": "N/A",
47
+ "model-ability": "Whether models can reason abductively about a given pair of observations."
48
+ },
49
+ "curation": {
50
+ "has-additional-curation": "no",
51
+ "modification-types": [],
52
+ "modification-description": "N/A",
53
+ "has-additional-splits": "no",
54
+ "additional-splits-description": "N/A",
55
+ "additional-splits-capacicites": "N/A"
56
+ },
57
+ "starting": {
58
+ "research-pointers": "- Paper: https://arxiv.org/abs/1908.05739\n- Code: https://github.com/allenai/abductive-commonsense-reasoning"
59
+ }
60
+ },
61
+ "results": {
62
+ "results": {
63
+ "model-abilities": "Whether models can reason abductively about a given pair of observations.",
64
+ "metrics": [
65
+ "BLEU",
66
+ "BERT-Score",
67
+ "ROUGE"
68
+ ],
69
+ "other-metrics-definitions": "N/A",
70
+ "has-previous-results": "no",
71
+ "current-evaluation": "N/A",
72
+ "previous-results": "N/A"
73
+ }
74
+ },
75
+ "curation": {
76
+ "original": {
77
+ "is-aggregated": "no",
78
+ "aggregated-sources": "N/A"
79
+ },
80
+ "language": {
81
+ "obtained": [
82
+ "Crowdsourced"
83
+ ],
84
+ "found": [],
85
+ "crowdsourced": [
86
+ "Amazon Mechanical Turk"
87
+ ],
88
+ "created": "N/A",
89
+ "machine-generated": "N/A",
90
+ "producers-description": "Language producers were English speakers in U.S., Canada, U.K and Australia.",
91
+ "topics": "No",
92
+ "validated": "validated by crowdworker",
93
+ "pre-processed": "N/A",
94
+ "is-filtered": "algorithmically",
95
+ "filtered-criteria": "Adversarial filtering algorithm as described in the paper: https://arxiv.org/abs/1908.05739"
96
+ },
97
+ "annotations": {
98
+ "origin": "automatically created",
99
+ "rater-number": "N/A",
100
+ "rater-qualifications": "N/A",
101
+ "rater-training-num": "N/A",
102
+ "rater-test-num": "N/A",
103
+ "rater-annotation-service-bool": "no",
104
+ "rater-annotation-service": [],
105
+ "values": "Each observation is associated with a list of COMET (https://arxiv.org/abs/1906.05317) inferences.",
106
+ "quality-control": "none",
107
+ "quality-control-details": "N/A"
108
+ },
109
+ "consent": {
110
+ "has-consent": "no",
111
+ "consent-policy": "N/A",
112
+ "consent-other": "N/A"
113
+ },
114
+ "pii": {
115
+ "has-pii": "no PII",
116
+ "no-pii-justification": "The dataset contains day-to-day events. It does not contain names, emails, addresses etc. ",
117
+ "pii-categories": [],
118
+ "is-pii-identified": "N/A",
119
+ "pii-identified-method": "N/A",
120
+ "is-pii-replaced": "N/A",
121
+ "pii-replaced-method": "N/A"
122
+ },
123
+ "maintenance": {
124
+ "has-maintenance": "no",
125
+ "description": "N/A",
126
+ "contact": "N/A",
127
+ "contestation-mechanism": "N/A",
128
+ "contestation-link": "N/A",
129
+ "contestation-description": "N/A"
130
+ }
131
+ },
132
+ "context": {
133
+ "previous": {
134
+ "is-deployed": "no",
135
+ "described-risks": "N/A",
136
+ "changes-from-observation": "N/A"
137
+ },
138
+ "underserved": {
139
+ "helps-underserved": "no",
140
+ "underserved-description": "N/A"
141
+ },
142
+ "biases": {
143
+ "has-biases": "no",
144
+ "bias-analyses": "N/A"
145
+ }
146
+ },
147
+ "considerations": {
148
+ "pii": {
149
+ "risks-description": "None"
150
+ },
151
+ "licenses": {
152
+ "dataset-restrictions": [
153
+ "public domain"
154
+ ],
155
+ "dataset-restrictions-other": "N/A",
156
+ "data-copyright": [
157
+ "public domain"
158
+ ],
159
+ "data-copyright-other": "N/A"
160
+ },
161
+ "limitations": {}
162
+ }
163
+ }