Datasets:
docs: add citation and arxiv link
Browse files- README.md +19 -4
- dataset_infos.json +1 -1
- prost.py +22 -7
README.md
CHANGED
@@ -54,7 +54,7 @@ task_ids:
|
|
54 |
|
55 |
- **Homepage:**
|
56 |
- **Repository:** https://github.com/nala-cub/prost
|
57 |
-
- **Paper:**
|
58 |
- **Leaderboard:**
|
59 |
- **Point of Contact:** [Stéphane Aroca-Ouellette](mailto:stephane.aroca-ouellette@colorado.edu)
|
60 |
|
@@ -158,12 +158,27 @@ PROST is designed to avoid models succeeding in unintended ways. First, PROST pr
|
|
158 |
|
159 |
### Licensing Information
|
160 |
|
161 |
-
|
162 |
|
163 |
### Citation Information
|
164 |
|
165 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
166 |
|
167 |
### Contributions
|
168 |
|
169 |
-
Thanks to [@
|
|
|
54 |
|
55 |
- **Homepage:**
|
56 |
- **Repository:** https://github.com/nala-cub/prost
|
57 |
+
- **Paper:** https://arxiv.org/abs/2106.03634
|
58 |
- **Leaderboard:**
|
59 |
- **Point of Contact:** [Stéphane Aroca-Ouellette](mailto:stephane.aroca-ouellette@colorado.edu)
|
60 |
|
|
|
158 |
|
159 |
### Licensing Information
|
160 |
|
161 |
+
PROST is licensed under the Apache 2.0 license.
|
162 |
|
163 |
### Citation Information
|
164 |
|
165 |
+
```
|
166 |
+
@inproceedings{aroca-ouellette-etal-2021-prost,
|
167 |
+
title = "{PROST}: {P}hysical Reasoning about Objects through Space and Time",
|
168 |
+
author = "Aroca-Ouellette, St{\'e}phane and
|
169 |
+
Paik, Cory and
|
170 |
+
Roncone, Alessandro and
|
171 |
+
Kann, Katharina",
|
172 |
+
booktitle = "Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021",
|
173 |
+
month = aug,
|
174 |
+
year = "2021",
|
175 |
+
address = "Online",
|
176 |
+
publisher = "Association for Computational Linguistics",
|
177 |
+
url = "https://aclanthology.org/2021.findings-acl.404",
|
178 |
+
pages = "4597--4608",
|
179 |
+
}
|
180 |
+
```
|
181 |
|
182 |
### Contributions
|
183 |
|
184 |
+
Thanks to [@corypaik](https://github.com/corypaik) for adding this dataset.
|
dataset_infos.json
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"default": {"description": "*Physical Reasoning about Objects Through Space and Time* (PROST) is a probing dataset to evaluate the ability of pretrained LMs to understand and reason about the physical world. PROST consists of 18,736 cloze-style multiple choice questions from 14 manually curated templates, covering 10 physical reasoning concepts: direction, mass, height, circumference, stackable, rollable, graspable, breakable, slideable, and bounceable.\n", "citation": "", "homepage": "https://github.com/nala-cub/prost", "license": "Apache 2.0", "features": {"A": {"dtype": "string", "id": null, "_type": "Value"}, "B": {"dtype": "string", "id": null, "_type": "Value"}, "C": {"dtype": "string", "id": null, "_type": "Value"}, "D": {"dtype": "string", "id": null, "_type": "Value"}, "context": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"dtype": "string", "id": null, "_type": "Value"}, "ex_question": {"dtype": "string", "id": null, "_type": "Value"}, "group": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"num_classes": 4, "names": ["A", "B", "C", "D"], "names_file": null, "id": null, "_type": "ClassLabel"}, "name": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "prost", "config_name": "default", "version": {"version_str": "1.0.
|
|
|
1 |
+
{"default": {"description": "*Physical Reasoning about Objects Through Space and Time* (PROST) is a probing dataset to evaluate the ability of pretrained LMs to understand and reason about the physical world. PROST consists of 18,736 cloze-style multiple choice questions from 14 manually curated templates, covering 10 physical reasoning concepts: direction, mass, height, circumference, stackable, rollable, graspable, breakable, slideable, and bounceable.\n", "citation": "@inproceedings{aroca-ouellette-etal-2021-prost,\n title = \"{PROST}: {P}hysical Reasoning about Objects through Space and Time\",\n author = \"Aroca-Ouellette, St{'e}phane and\n Paik, Cory and\n Roncone, Alessandro and\n Kann, Katharina\",\n booktitle = \"Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021\",\n month = aug,\n year = \"2021\",\n address = \"Online\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://aclanthology.org/2021.findings-acl.404\",\n pages = \"4597--4608\",\n}\n", "homepage": "https://github.com/nala-cub/prost", "license": "Apache 2.0", "features": {"A": {"dtype": "string", "id": null, "_type": "Value"}, "B": {"dtype": "string", "id": null, "_type": "Value"}, "C": {"dtype": "string", "id": null, "_type": "Value"}, "D": {"dtype": "string", "id": null, "_type": "Value"}, "context": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"dtype": "string", "id": null, "_type": "Value"}, "ex_question": {"dtype": "string", "id": null, "_type": "Value"}, "group": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"num_classes": 4, "names": ["A", "B", "C", "D"], "names_file": null, "id": null, "_type": "ClassLabel"}, "name": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "prost", "config_name": "default", "version": {"version_str": "1.0.1", "description": null, "major": 1, "minor": 0, "patch": 1}, "splits": {"test": {"name": "test", "num_bytes": 5698256, "num_examples": 18736, "dataset_name": "prost"}}, "download_checksums": {"https://huggingface.co/datasets/corypaik/prost/resolve/main/data/default.jsonl": {"num_bytes": 7197063, "checksum": "05d773ae2768047579ea22415fc593199430747027b3df5621e60044fe893bde"}}, "download_size": 7197063, "post_processing_size": null, "dataset_size": 5698256, "size_in_bytes": 12895319}}
|
prost.py
CHANGED
@@ -12,17 +12,33 @@
|
|
12 |
# See the License for the specific language governing permissions and
|
13 |
# limitations under the License.
|
14 |
# ==============================================================================
|
15 |
-
"""
|
16 |
-
|
|
|
|
|
|
|
17 |
import json
|
18 |
import datasets
|
19 |
|
20 |
-
|
21 |
-
# Find for instance the citation on arxiv or on the dataset repo/website
|
22 |
_CITATION = """\
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
23 |
"""
|
24 |
|
25 |
-
|
26 |
_DESCRIPTION = """\
|
27 |
*Physical Reasoning about Objects Through Space and Time* (PROST) is a probing dataset to evaluate the ability of pretrained LMs to understand and reason about the physical world. PROST consists of 18,736 cloze-style multiple choice questions from 14 manually curated templates, covering 10 physical reasoning concepts: direction, mass, height, circumference, stackable, rollable, graspable, breakable, slideable, and bounceable.
|
28 |
"""
|
@@ -38,9 +54,8 @@ MC_LABELS = list('ABCD')
|
|
38 |
|
39 |
|
40 |
class Prost(datasets.GeneratorBasedBuilder):
|
41 |
-
"""TODO: Short description of my dataset."""
|
42 |
|
43 |
-
VERSION = datasets.Version('1.0.
|
44 |
|
45 |
def _info(self):
|
46 |
features = datasets.Features({
|
|
|
12 |
# See the License for the specific language governing permissions and
|
13 |
# limitations under the License.
|
14 |
# ==============================================================================
|
15 |
+
""" Physical Reasoning about Objects Through Space and Time (PROST)
|
16 |
+
|
17 |
+
PROST is a probing dataset to evaluate the ability of pretrained LMs to
|
18 |
+
understand and reason about the physical world.
|
19 |
+
"""
|
20 |
import json
|
21 |
import datasets
|
22 |
|
23 |
+
|
|
|
24 |
_CITATION = """\
|
25 |
+
@inproceedings{aroca-ouellette-etal-2021-prost,
|
26 |
+
title = "{PROST}: {P}hysical Reasoning about Objects through Space and Time",
|
27 |
+
author = "Aroca-Ouellette, St{\'e}phane and
|
28 |
+
Paik, Cory and
|
29 |
+
Roncone, Alessandro and
|
30 |
+
Kann, Katharina",
|
31 |
+
booktitle = "Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021",
|
32 |
+
month = aug,
|
33 |
+
year = "2021",
|
34 |
+
address = "Online",
|
35 |
+
publisher = "Association for Computational Linguistics",
|
36 |
+
url = "https://aclanthology.org/2021.findings-acl.404",
|
37 |
+
pages = "4597--4608",
|
38 |
+
}
|
39 |
"""
|
40 |
|
41 |
+
|
42 |
_DESCRIPTION = """\
|
43 |
*Physical Reasoning about Objects Through Space and Time* (PROST) is a probing dataset to evaluate the ability of pretrained LMs to understand and reason about the physical world. PROST consists of 18,736 cloze-style multiple choice questions from 14 manually curated templates, covering 10 physical reasoning concepts: direction, mass, height, circumference, stackable, rollable, graspable, breakable, slideable, and bounceable.
|
44 |
"""
|
|
|
54 |
|
55 |
|
56 |
class Prost(datasets.GeneratorBasedBuilder):
|
|
|
57 |
|
58 |
+
VERSION = datasets.Version('1.0.1')
|
59 |
|
60 |
def _info(self):
|
61 |
features = datasets.Features({
|