parquet-converter commited on
Commit
d1ce886
1 Parent(s): cca3cfb

Update parquet files

Browse files
Files changed (49) hide show
  1. .gitattributes +11 -0
  2. README.md +0 -69
  3. dataset_infos.json +0 -1
  4. xlwic.py +0 -231
  5. xlwic_de_de/xlwic-test.parquet +3 -0
  6. xlwic_de_de/xlwic-train.parquet +3 -0
  7. xlwic_de_de/xlwic-validation.parquet +3 -0
  8. xlwic_en_bg/xlwic-test.parquet +0 -0
  9. xlwic_en_bg/xlwic-train.parquet +0 -0
  10. xlwic_en_bg/xlwic-validation.parquet +0 -0
  11. xlwic_en_da/xlwic-test.parquet +0 -0
  12. xlwic_en_da/xlwic-train.parquet +0 -0
  13. xlwic_en_da/xlwic-validation.parquet +0 -0
  14. xlwic_en_de/xlwic-test.parquet +3 -0
  15. xlwic_en_de/xlwic-train.parquet +0 -0
  16. xlwic_en_de/xlwic-validation.parquet +3 -0
  17. xlwic_en_et/xlwic-test.parquet +0 -0
  18. xlwic_en_et/xlwic-train.parquet +0 -0
  19. xlwic_en_et/xlwic-validation.parquet +0 -0
  20. xlwic_en_fa/xlwic-test.parquet +0 -0
  21. xlwic_en_fa/xlwic-train.parquet +0 -0
  22. xlwic_en_fa/xlwic-validation.parquet +0 -0
  23. xlwic_en_fr/xlwic-test.parquet +3 -0
  24. xlwic_en_fr/xlwic-train.parquet +0 -0
  25. xlwic_en_fr/xlwic-validation.parquet +3 -0
  26. xlwic_en_hr/xlwic-test.parquet +0 -0
  27. xlwic_en_hr/xlwic-train.parquet +0 -0
  28. xlwic_en_hr/xlwic-validation.parquet +0 -0
  29. xlwic_en_it/xlwic-test.parquet +0 -0
  30. xlwic_en_it/xlwic-train.parquet +0 -0
  31. xlwic_en_it/xlwic-validation.parquet +0 -0
  32. xlwic_en_ja/xlwic-test.parquet +0 -0
  33. xlwic_en_ja/xlwic-train.parquet +0 -0
  34. xlwic_en_ja/xlwic-validation.parquet +0 -0
  35. xlwic_en_ko/xlwic-test.parquet +0 -0
  36. xlwic_en_ko/xlwic-train.parquet +0 -0
  37. xlwic_en_ko/xlwic-validation.parquet +0 -0
  38. xlwic_en_nl/xlwic-test.parquet +0 -0
  39. xlwic_en_nl/xlwic-train.parquet +0 -0
  40. xlwic_en_nl/xlwic-validation.parquet +0 -0
  41. xlwic_en_zh/xlwic-test.parquet +3 -0
  42. xlwic_en_zh/xlwic-train.parquet +0 -0
  43. xlwic_en_zh/xlwic-validation.parquet +0 -0
  44. xlwic_fr_fr/xlwic-test.parquet +3 -0
  45. xlwic_fr_fr/xlwic-train.parquet +3 -0
  46. xlwic_fr_fr/xlwic-validation.parquet +3 -0
  47. xlwic_it_it/xlwic-test.parquet +0 -0
  48. xlwic_it_it/xlwic-train.parquet +0 -0
  49. xlwic_it_it/xlwic-validation.parquet +0 -0
.gitattributes CHANGED
@@ -14,3 +14,14 @@
14
  *.pb filter=lfs diff=lfs merge=lfs -text
15
  *.pt filter=lfs diff=lfs merge=lfs -text
16
  *.pth filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
14
  *.pb filter=lfs diff=lfs merge=lfs -text
15
  *.pt filter=lfs diff=lfs merge=lfs -text
16
  *.pth filter=lfs diff=lfs merge=lfs -text
17
+ xlwic_de_de/xlwic-train.parquet filter=lfs diff=lfs merge=lfs -text
18
+ xlwic_de_de/xlwic-validation.parquet filter=lfs diff=lfs merge=lfs -text
19
+ xlwic_de_de/xlwic-test.parquet filter=lfs diff=lfs merge=lfs -text
20
+ xlwic_en_de/xlwic-validation.parquet filter=lfs diff=lfs merge=lfs -text
21
+ xlwic_en_de/xlwic-test.parquet filter=lfs diff=lfs merge=lfs -text
22
+ xlwic_en_fr/xlwic-validation.parquet filter=lfs diff=lfs merge=lfs -text
23
+ xlwic_en_fr/xlwic-test.parquet filter=lfs diff=lfs merge=lfs -text
24
+ xlwic_en_zh/xlwic-test.parquet filter=lfs diff=lfs merge=lfs -text
25
+ xlwic_fr_fr/xlwic-train.parquet filter=lfs diff=lfs merge=lfs -text
26
+ xlwic_fr_fr/xlwic-validation.parquet filter=lfs diff=lfs merge=lfs -text
27
+ xlwic_fr_fr/xlwic-test.parquet filter=lfs diff=lfs merge=lfs -text
README.md DELETED
@@ -1,69 +0,0 @@
1
- ---
2
- annotations_creators:
3
- - expert-generated
4
- extended:
5
- - original
6
- language_creators:
7
- - found
8
- language:
9
- - en
10
- - bg
11
- - zh
12
- - hr
13
- - da
14
- - nl
15
- - et
16
- - fa
17
- - ja
18
- - ko
19
- - it
20
- - fr
21
- - de
22
- license:
23
- - cc-by-nc-4.0
24
- multilinguality:
25
- - multilingual
26
- size_categories:
27
- - 10K<n<100K
28
- source_datasets:
29
- - original
30
- task_categories:
31
- - text-classification
32
- task_ids:
33
- - semantic-similarity-classification
34
- ---
35
-
36
- # XL-WiC
37
- Huggingface dataset for the XL-WiC paper [https://www.aclweb.org/anthology/2020.emnlp-main.584.pdf](https://www.aclweb.org/anthology/2020.emnlp-main.584.pdf).
38
- Please refer to the official [website](https://pilehvar.github.io/xlwic/) for more information.
39
-
40
-
41
- ## Configurations
42
- When loading one of the XL-WSD datasets one has to specify the training language and the target language (on which dev and test will be performed).
43
- Please refer to [Languages](#languages) section to see in which languages training data is available.
44
- For example, we can load the dataset having English as training language and Italian as target language as follows:
45
- ```python
46
- from datasets import load_dataset
47
- dataset = load_dataset('pasinit/xlwic', 'en_it')
48
- ```
49
-
50
- ## Languages
51
- **Training data**
52
- - en (English)
53
- - fr (French)
54
- - de (German)
55
- - it (Italian)
56
-
57
- **Dev & Test data**
58
- - fr (French)
59
- - de (German)
60
- - it (Italian)
61
- - bg (Bulgarian)
62
- - zh (Chinese)
63
- - hr (Croatian)
64
- - da (Danish)
65
- - nl (Dutch)
66
- - et (Estonian)
67
- - fa (Farsi)
68
- - ja (Japanesse)
69
- - ko (Korean)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
dataset_infos.json DELETED
@@ -1 +0,0 @@
1
- {"xlwic_en_bg": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_bg", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 139338, "num_examples": 998, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 173903, "num_examples": 1220, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 1046528, "size_in_bytes": 19794686}, "xlwic_en_zh": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_zh", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 779740, "num_examples": 3046, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 1434016, "num_examples": 5538, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 2947043, "size_in_bytes": 21695201}, "xlwic_en_hr": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_hr", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 16877, "num_examples": 104, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 66694, "num_examples": 408, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 816858, "size_in_bytes": 19565016}, "xlwic_en_da": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_da", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 198876, "num_examples": 852, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 800224, "num_examples": 3406, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 1732387, "size_in_bytes": 20480545}, "xlwic_en_nl": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_nl", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 40702, "num_examples": 250, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 163964, "num_examples": 1004, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 937953, "size_in_bytes": 19686111}, "xlwic_en_et": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_et", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 16685, "num_examples": 98, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 67034, "num_examples": 390, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 817006, "size_in_bytes": 19565164}, "xlwic_en_fa": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_fa", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 54430, "num_examples": 200, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 208448, "num_examples": 800, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 996165, "size_in_bytes": 19744323}, "xlwic_en_ja": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_ja", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 32438, "num_examples": 208, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 127556, "num_examples": 824, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 893281, "size_in_bytes": 19641439}, "xlwic_en_ko": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_ko", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 42088, "num_examples": 404, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 110032, "num_examples": 1014, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 885407, "size_in_bytes": 19633565}, "xlwic_en_it": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_it", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 35720, "num_examples": 198, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 109903, "num_examples": 592, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 878910, "size_in_bytes": 19627068}, "xlwic_en_fr": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_fr", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 2045677, "num_examples": 8588, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 5615634, "num_examples": 22232, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 8394598, "size_in_bytes": 27142756}, "xlwic_en_de": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_en_de", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 733287, "num_examples": 5428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 2277398, "num_examples": 8870, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 6360304, "num_examples": 24268, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 9370989, "size_in_bytes": 28119147}, "xlwic_it_it": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_it_it", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 204644, "num_examples": 1144, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 35720, "num_examples": 198, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 109903, "num_examples": 592, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 350267, "size_in_bytes": 19098425}, "xlwic_fr_fr": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_fr_fr", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 10210148, "num_examples": 39428, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 2045677, "num_examples": 8588, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 5615634, "num_examples": 22232, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 17871459, "size_in_bytes": 36619617}, "xlwic_de_de": {"description": "A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.\n\nXL-WiC provides dev and test sets in the following 12 languages:\n\nBulgarian (BG)\nDanish (DA)\nGerman (DE)\nEstonian (ET)\nFarsi (FA)\nFrench (FR)\nCroatian (HR)\nItalian (IT)\nJapanese (JA)\nKorean (KO)\nDutch (NL)\nChinese (ZH)\nand training sets in the following 3 languages:\n\nGerman (DE)\nFrench (FR)\nItalian (IT)\n", "citation": "@inproceedings{raganato-etal-2020-xl-wic,\n title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},\n author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},\n booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},\n pages={7193--7206},\n year={2020}\n}\n", "homepage": "https://pilehvar.github.io/xlwic/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "context_1": {"dtype": "string", "id": null, "_type": "Value"}, "context_2": {"dtype": "string", "id": null, "_type": "Value"}, "target_word": {"dtype": "string", "id": null, "_type": "Value"}, "pos": {"dtype": "string", "id": null, "_type": "Value"}, "target_word_location_1": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "target_word_location_2": {"char_start": {"dtype": "int32", "id": null, "_type": "Value"}, "char_end": {"dtype": "int32", "id": null, "_type": "Value"}}, "language": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "int32", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "xlwic", "config_name": "xlwic_de_de", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 12625723, "num_examples": 48042, "dataset_name": "xlwic"}, "validation": {"name": "validation", "num_bytes": 2277398, "num_examples": 8870, "dataset_name": "xlwic"}, "test": {"name": "test", "num_bytes": 6360304, "num_examples": 24268, "dataset_name": "xlwic"}}, "download_checksums": {"https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip": {"num_bytes": 18748158, "checksum": "40d12d338c90937eab820f24206c06d58b21d996b1d5c7bf69b3f8c65cf7eaff"}}, "download_size": 18748158, "post_processing_size": null, "dataset_size": 21263425, "size_in_bytes": 40011583}}
 
 
xlwic.py DELETED
@@ -1,231 +0,0 @@
1
- from dataclasses import dataclass
2
- import datasets
3
- from datasets.info import DatasetInfo
4
- from datasets.utils.download_manager import DownloadManager
5
- import os
6
-
7
- _DESCRIPTION = """A system's task on any of the XL-WiC datasets is to identify the intended meaning of a word in a context of a given language. XL-WiC is framed as a binary classification task. Each instance in XL-WiC has a target word w, either a verb or a noun, for which two contexts are provided. Each of these contexts triggers a specific meaning of w. The task is to identify if the occurrences of w in the two contexts correspond to the same meaning or not.
8
-
9
- XL-WiC provides dev and test sets in the following 12 languages:
10
-
11
- Bulgarian (BG)
12
- Danish (DA)
13
- German (DE)
14
- Estonian (ET)
15
- Farsi (FA)
16
- French (FR)
17
- Croatian (HR)
18
- Italian (IT)
19
- Japanese (JA)
20
- Korean (KO)
21
- Dutch (NL)
22
- Chinese (ZH)
23
- and training sets in the following 3 languages:
24
-
25
- German (DE)
26
- French (FR)
27
- Italian (IT)
28
- """
29
- _CITATION = """@inproceedings{raganato-etal-2020-xl-wic,
30
- title={XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization},
31
- author={Raganato, Alessandro and Pasini, Tommaso and Camacho-Collados, Jose and Pilehvar, Mohammad Taher},
32
- booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
33
- pages={7193--7206},
34
- year={2020}
35
- }
36
- """
37
- _DOWNLOAD_URL = "https://pilehvar.github.io/xlwic/data/xlwic_datasets.zip"
38
- _VERSION = "1.0.0"
39
- _WN_LANGS = ["EN", "BG", "ZH", "HR", "DA", "NL", "ET", "FA", "JA", "KO"]
40
- _WIKT_LANGS = ["IT", "FR", "DE"]
41
- _CODE_TO_LANG_ID = {
42
- "EN": "english",
43
- "BG": "bulgarian",
44
- "ZH": "chinese",
45
- "HR": "croatian",
46
- "DA": "danish",
47
- "NL": "dutch",
48
- "ET": "estonian",
49
- "FA": "farsi",
50
- "JA": "japanese",
51
- "KO": "korean",
52
- "IT": "italian",
53
- "FR": "french",
54
- "DE": "german",
55
- }
56
- _AVAILABLE_PAIRS = (
57
- list(zip(["EN"] * (len(_WN_LANGS) - 1), _WN_LANGS[1:]))
58
- + list(zip(["EN"] * len(_WIKT_LANGS), _WIKT_LANGS))
59
- + [("IT", "IT"), ("FR", "FR"), ("DE", "DE")]
60
- )
61
-
62
- @dataclass
63
- class XLWiCConfig(datasets.BuilderConfig):
64
- version:str=None
65
- training_lang:str = None
66
- target_lang:str = None
67
- name:str = None
68
-
69
-
70
- class XLWIC(datasets.GeneratorBasedBuilder):
71
- BUILDER_CONFIGS = [
72
- XLWiCConfig(
73
- name=f"xlwic_{source.lower()}_{target.lower()}",
74
- training_lang=source,
75
- target_lang=target,
76
- version=datasets.Version(_VERSION, ""),
77
- )
78
- for source, target in _AVAILABLE_PAIRS
79
- ]
80
-
81
- def _info(self) -> DatasetInfo:
82
- return datasets.DatasetInfo(
83
- description=_DESCRIPTION,
84
- features=datasets.Features(
85
- {
86
- "id": datasets.Value("string"),
87
- "context_1": datasets.Value("string"),
88
- "context_2": datasets.Value("string"),
89
- "target_word": datasets.Value("string"),
90
- "pos": datasets.Value("string"),
91
- "target_word_location_1":
92
- {
93
- "char_start": datasets.Value("int32"),
94
- "char_end": datasets.Value("int32"),
95
- },
96
- "target_word_location_2":
97
- {
98
- "char_start": datasets.Value("int32"),
99
- "char_end": datasets.Value("int32"),
100
- },
101
- "language": datasets.Value("string"),
102
- "label": datasets.Value("int32"),
103
- }
104
- ),
105
- supervised_keys=None,
106
- homepage="https://pilehvar.github.io/xlwic/",
107
- citation=_CITATION,
108
- )
109
-
110
- def _split_generators(self, dl_manager: DownloadManager):
111
- downloaded_file = dl_manager.download_and_extract(_DOWNLOAD_URL)
112
- dataset_root_folder = os.path.join(downloaded_file, "xlwic_datasets")
113
-
114
- return [
115
- datasets.SplitGenerator(
116
- name=datasets.Split.TRAIN,
117
- # These kwargs will be passed to _generate_examples
118
- gen_kwargs={
119
- "dataset_root": dataset_root_folder,
120
- "lang": self.config.training_lang,
121
- "split": "train",
122
- },
123
- ),
124
- datasets.SplitGenerator(
125
- name=datasets.Split.VALIDATION,
126
- # These kwargs will be passed to _generate_examples
127
- gen_kwargs={
128
- "dataset_root": dataset_root_folder,
129
- "lang": self.config.target_lang,
130
- "split": "valid",
131
- },
132
- ),
133
- datasets.SplitGenerator(
134
- name=datasets.Split.TEST,
135
- # These kwargs will be passed to _generate_examples
136
- gen_kwargs={
137
- "dataset_root": dataset_root_folder,
138
- "lang": self.config.target_lang,
139
- "split": "test",
140
- },
141
- ),
142
- ]
143
-
144
- def _yield_from_lines(self, lines, lang):
145
-
146
- for i, (
147
- tw,
148
- pos,
149
- char_start_1,
150
- char_end_1,
151
- char_start_2,
152
- char_end_2,
153
- context_1,
154
- context_2,
155
- label,
156
- ) in enumerate(lines):
157
- _id = f"{lang}_{i}"
158
- yield _id, {
159
- "id": _id,
160
- "target_word": tw,
161
- "context_1": context_1,
162
- "context_2": context_2,
163
- "label": int(label),
164
- "target_word_location_1": {
165
- "char_start": int(char_start_1),
166
- "char_end": int(char_end_1),
167
- },
168
- "target_word_location_2": {
169
- "char_start": int(char_start_2),
170
- "char_end": int(char_end_2)
171
- },
172
- "pos": pos,
173
- "language": lang,
174
- }
175
-
176
- def _from_selfcontained_file(self, dataset_root, lang, split):
177
- ext_lang = _CODE_TO_LANG_ID[lang]
178
- if lang in _WIKT_LANGS:
179
- path = os.path.join(
180
- dataset_root,
181
- "xlwic_wikt",
182
- f"{ext_lang}_{lang.lower()}",
183
- f"{lang.lower()}_{split}.txt",
184
- )
185
- elif lang != "EN" and lang in _WN_LANGS:
186
- path = os.path.join(
187
- dataset_root,
188
- "xlwic_wn",
189
- f"{ext_lang}_{lang.lower()}",
190
- f"{lang.lower()}_{split}.txt",
191
- )
192
- elif lang == "EN" and lang in _WN_LANGS:
193
- path = os.path.join(
194
- dataset_root, "wic_english", f"{split}_{lang.lower()}.txt"
195
- )
196
- with open(path) as lines:
197
- all_lines = [line.strip().split("\t") for line in lines]
198
- yield from self._yield_from_lines(all_lines, lang)
199
-
200
- def _from_test_files(self, dataset_root, lang, split):
201
- ext_lang = _CODE_TO_LANG_ID[lang]
202
- if lang in _WIKT_LANGS:
203
- path_data = os.path.join(
204
- dataset_root,
205
- "xlwic_wikt",
206
- f"{ext_lang}_{lang.lower()}",
207
- f"{lang.lower()}_{split}_data.txt",
208
- )
209
- elif lang != "EN" and lang in _WN_LANGS:
210
- path_data = os.path.join(
211
- dataset_root,
212
- "xlwic_wn",
213
- f"{ext_lang}_{lang.lower()}",
214
- f"{lang.lower()}_{split}_data.txt",
215
- )
216
- path_gold = path_data.replace('_data.txt', '_gold.txt')
217
- with open(path_data) as lines:
218
- all_lines = [line.strip().split("\t") for line in lines]
219
- with open(path_gold) as lines:
220
- all_labels = [line.strip() for line in lines]
221
- for line, label in zip(all_lines, all_labels):
222
- line.append(label)
223
- yield from self._yield_from_lines(all_lines, lang)
224
-
225
-
226
- def _generate_examples(self, dataset_root, lang, split, **kwargs):
227
- if split in {"train", "valid"}:
228
- yield from self._from_selfcontained_file(dataset_root, lang, split)
229
- else:
230
- yield from self._from_test_files(dataset_root, lang, split)
231
-
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
xlwic_de_de/xlwic-test.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ec6270249bf18af487f95531453e410125c039ac2f400577082ab7c1de7c3de
3
+ size 4296957
xlwic_de_de/xlwic-train.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:edc44e1695c75c0a1a0e348eedd51e017106b0a0e00a12d8f7c0a97b96e6041d
3
+ size 8495279
xlwic_de_de/xlwic-validation.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fd02b2d1681e99c35386770ef91949905c2eb21e24413eca791f9157ba81cfe
3
+ size 1444821
xlwic_en_bg/xlwic-test.parquet ADDED
Binary file (82 kB). View file
 
xlwic_en_bg/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_bg/xlwic-validation.parquet ADDED
Binary file (59.2 kB). View file
 
xlwic_en_da/xlwic-test.parquet ADDED
Binary file (533 kB). View file
 
xlwic_en_da/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_da/xlwic-validation.parquet ADDED
Binary file (128 kB). View file
 
xlwic_en_de/xlwic-test.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ec6270249bf18af487f95531453e410125c039ac2f400577082ab7c1de7c3de
3
+ size 4296957
xlwic_en_de/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_de/xlwic-validation.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fd02b2d1681e99c35386770ef91949905c2eb21e24413eca791f9157ba81cfe
3
+ size 1444821
xlwic_en_et/xlwic-test.parquet ADDED
Binary file (49.2 kB). View file
 
xlwic_en_et/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_et/xlwic-validation.parquet ADDED
Binary file (18 kB). View file
 
xlwic_en_fa/xlwic-test.parquet ADDED
Binary file (105 kB). View file
 
xlwic_en_fa/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_fa/xlwic-validation.parquet ADDED
Binary file (35.4 kB). View file
 
xlwic_en_fr/xlwic-test.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0855494b8c2c18d6051abaf4de5e9176eea8207a2dbeac4c46d171fc20b7a50
3
+ size 3737736
xlwic_en_fr/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_fr/xlwic-validation.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:190aec5760177fe92a1a270c066bfd7dc067005285e46da972e1b4811baac204
3
+ size 1269199
xlwic_en_hr/xlwic-test.parquet ADDED
Binary file (48.2 kB). View file
 
xlwic_en_hr/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_hr/xlwic-validation.parquet ADDED
Binary file (18.3 kB). View file
 
xlwic_en_it/xlwic-test.parquet ADDED
Binary file (69.9 kB). View file
 
xlwic_en_it/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_it/xlwic-validation.parquet ADDED
Binary file (28.9 kB). View file
 
xlwic_en_ja/xlwic-test.parquet ADDED
Binary file (68.6 kB). View file
 
xlwic_en_ja/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_ja/xlwic-validation.parquet ADDED
Binary file (23.4 kB). View file
 
xlwic_en_ko/xlwic-test.parquet ADDED
Binary file (57.6 kB). View file
 
xlwic_en_ko/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_ko/xlwic-validation.parquet ADDED
Binary file (25.2 kB). View file
 
xlwic_en_nl/xlwic-test.parquet ADDED
Binary file (92.9 kB). View file
 
xlwic_en_nl/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_nl/xlwic-validation.parquet ADDED
Binary file (29.2 kB). View file
 
xlwic_en_zh/xlwic-test.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a90464080f7690fd77b5938dc1aa96b681e500e0fb9b68334c01499b2283bc1d
3
+ size 1056045
xlwic_en_zh/xlwic-train.parquet ADDED
Binary file (397 kB). View file
 
xlwic_en_zh/xlwic-validation.parquet ADDED
Binary file (556 kB). View file
 
xlwic_fr_fr/xlwic-test.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0855494b8c2c18d6051abaf4de5e9176eea8207a2dbeac4c46d171fc20b7a50
3
+ size 3737736
xlwic_fr_fr/xlwic-train.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:736b4216da3b2f340e3cf621e7f51ba3d47b3cde7fec247b6ad9e2e5ac200f11
3
+ size 6790869
xlwic_fr_fr/xlwic-validation.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:190aec5760177fe92a1a270c066bfd7dc067005285e46da972e1b4811baac204
3
+ size 1269199
xlwic_it_it/xlwic-test.parquet ADDED
Binary file (69.9 kB). View file
 
xlwic_it_it/xlwic-train.parquet ADDED
Binary file (126 kB). View file
 
xlwic_it_it/xlwic-validation.parquet ADDED
Binary file (28.9 kB). View file