mbayan commited on
Commit
1888764
·
verified ·
1 Parent(s): f31866f

Update dataset card

Browse files
Files changed (1) hide show
  1. README.md +37 -50
README.md CHANGED
@@ -26,7 +26,9 @@ configs:
26
  path: data/dev_test-*
27
  - split: test
28
  path: data/test-*
 
29
  dataset_info:
 
30
  features:
31
  - name: id
32
  dtype: string
@@ -40,19 +42,13 @@ dataset_info:
40
  sequence: string
41
  splits:
42
  - name: train
43
- num_bytes: 274812860.5
44
  num_examples: 3500
45
  - name: dev
46
- num_bytes: 40130069.0
47
  num_examples: 500
48
  - name: dev_test
49
- num_bytes: 38915856.0
50
  num_examples: 500
51
  - name: test
52
- num_bytes: 38669448.0
53
  num_examples: 500
54
- download_size: 386298290
55
- dataset_size: 392528233.5
56
  ---
57
 
58
  # ArGuard – Track A: Arabic Hateful Memes
@@ -81,30 +77,33 @@ Given a meme (image + Arabic text):
81
 
82
  ## Splits
83
 
84
- | Split | Records | Labels | Source | Released |
85
- |--------------|---------|--------------|----------------------------------------|---------------------------|
86
- | `train` | 3,500 | full | single-annotated bulk | development phase |
87
- | `dev` | 500 | full | single-annotated bulk | development phase |
88
- | `dev_test` | 500 | **dropped** | single-annotated test sample | development phase (leaderboard) |
89
- | `test` | 500 | full | **triple-annotated gold** (calibration) | final-evaluation phase |
90
-
91
- - `dev_test` is the **leaderboard set** for the development phase. Labels
92
- are intentionally stripped (`label = null`, `fine_grained_label = []`)
93
- and will be released only after the development phase closes.
94
- - `test` is the **held-out blind test** for final ranking. All 500 records
95
- are triple-annotated with majority voting. This split is not part of
96
- the public release and will appear here only when the final-evaluation
97
- phase begins.
 
98
 
99
  ### Binary label distribution
100
 
101
  | Split | Hateful | Not Hateful | % Hateful |
102
  |------------|--------:|------------:|----------:|
103
  | train | 1,324 | 2,176 | 37.8% |
104
- | dev | 189 | 311 | 37.8% |
105
- | dev_test | 189 | 311 | 37.8% |
106
- | test | 148 | 352 | 29.6% |
107
- | **Total** | **1,850** | **3,150** | 37.0% |
 
 
108
 
109
  ### Fine-grained sub-types (Subtask A2)
110
 
@@ -134,8 +133,8 @@ than the meme counts.
134
  "id": "f9a8…b1.jpg", # str – original image filename, unique
135
  "image": <PIL.Image.Image>, # embedded bytes, decoded on access
136
  "text": "…", # str – OCR-extracted Arabic meme text
137
- "label": "Hateful" | "Not Hateful" | None, # None on dev_test
138
- "fine_grained_label": [...], # list[str] – empty on dev_test
139
  }
140
  ```
141
 
@@ -151,14 +150,20 @@ train_ex = ds["train"][0]
151
  train_ex["image"].show()
152
  print(train_ex["text"], train_ex["label"], train_ex["fine_grained_label"])
153
 
154
- # dev_test is unlabelled — used only to produce leaderboard submissions
155
- print(ds["dev_test"][0]["label"]) # -> None
156
  ```
157
 
 
 
 
 
158
  ## Shared-task resources
159
 
160
- - **Website:** https://araieval.gitlab.io/ArGuard2026/
161
  - **Starter kit / baselines / scorers (GitHub):** https://github.com/araieval/ArGuard-2026-tasks
 
 
162
  - **Contact organisers:** arguard2026-organizers@googlegroups.com
163
 
164
  ## Annotation
@@ -191,29 +196,11 @@ generate harmful content.
191
 
192
  ## Citation
193
 
194
- ```bibtex
195
- @inproceedings{alam-etal-2026-arguard,
196
- title = {{ArGuard Shared Task}: Harmful Content Detection in
197
- {A}rabic {M}emes and {LLM} {P}rompts},
198
- author = {Alam, Firoj and
199
- Biswas, Md. Rafiul and
200
- Kmainasi, Mohamed Bayan and
201
- Shahroor, Ali Ezzat and
202
- Mubarak, Hamdy and
203
- Mikros, Georgios and
204
- Hasnat, Abul and
205
- Zaghouani, Wajdi},
206
- booktitle = {Proceedings of the Fourth Arabic Natural Language
207
- Processing Conference: Shared Tasks},
208
- month = oct,
209
- year = {2026},
210
- address = {Budapest, Hungary},
211
- publisher = {Association for Computational Linguistics}
212
- }
213
- ```
214
 
215
  ## Contact
216
 
217
  - **Email:** arguard2026-organizers@googlegroups.com
218
- - **Website:** https://araieval.gitlab.io/ArGuard2026/
219
  - **GitHub:** https://github.com/araieval/ArGuard-2026-tasks
 
26
  path: data/dev_test-*
27
  - split: test
28
  path: data/test-*
29
+ default: true
30
  dataset_info:
31
+ config_name: default
32
  features:
33
  - name: id
34
  dtype: string
 
42
  sequence: string
43
  splits:
44
  - name: train
 
45
  num_examples: 3500
46
  - name: dev
 
47
  num_examples: 500
48
  - name: dev_test
 
49
  num_examples: 500
50
  - name: test
 
51
  num_examples: 500
 
 
52
  ---
53
 
54
  # ArGuard – Track A: Arabic Hateful Memes
 
77
 
78
  ## Splits
79
 
80
+ | Split | Records | Labels | Source | Released |
81
+ |--------------|---------|--------------|-----------------------------------------|---------------------------------|
82
+ | `train` | 3,500 | full | single-annotated bulk | development phase |
83
+ | `dev` | 500 | full | single-annotated bulk | development phase |
84
+ | `dev_test` | 500 | **dropped** | single-annotated test sample | development phase (leaderboard) |
85
+ | `test` | 500 | **dropped** | **triple-annotated gold** (calibration) | final-evaluation phase (leaderboard) |
86
+
87
+ - `dev_test` was the **leaderboard set** for the development phase, which
88
+ is now closed. Labels are stripped (`label = null`,
89
+ `fine_grained_label = []`).
90
+ - `test` is the **blind test set** for the final-evaluation phase and is
91
+ the file you submit predictions for. It is released the same way as
92
+ `dev_test` — images and OCR text only, labels stripped. All 500 records
93
+ are triple-annotated with majority voting; the gold labels are held in
94
+ the CodaBench reference data and are never published here.
95
 
96
  ### Binary label distribution
97
 
98
  | Split | Hateful | Not Hateful | % Hateful |
99
  |------------|--------:|------------:|----------:|
100
  | train | 1,324 | 2,176 | 37.8% |
101
+ | dev | 189 | 311 | 37.8% |
102
+ | dev_test | 189 | 311 | 37.8% |
103
+ | **Total** | **1,702** | **2,798** | 37.8% |
104
+
105
+ The label distribution of the blind `test` split is withheld during the
106
+ final-evaluation phase.
107
 
108
  ### Fine-grained sub-types (Subtask A2)
109
 
 
133
  "id": "f9a8…b1.jpg", # str – original image filename, unique
134
  "image": <PIL.Image.Image>, # embedded bytes, decoded on access
135
  "text": "…", # str – OCR-extracted Arabic meme text
136
+ "label": "Hateful" | "Not Hateful" | None, # None on dev_test / test
137
+ "fine_grained_label": [...], # list[str] – empty on dev_test / test
138
  }
139
  ```
140
 
 
150
  train_ex["image"].show()
151
  print(train_ex["text"], train_ex["label"], train_ex["fine_grained_label"])
152
 
153
+ # dev_test and test are unlabelled — used only to produce submissions
154
+ print(ds["test"][0]["label"]) # -> None
155
  ```
156
 
157
+ The blind `test` split is the submission target for the final-evaluation
158
+ phase. Train on `train` + `dev`, predict on `test`, and upload to
159
+ CodaBench (links below).
160
+
161
  ## Shared-task resources
162
 
163
+ - **Website:** https://araieval.github.io/ArGuard2026/
164
  - **Starter kit / baselines / scorers (GitHub):** https://github.com/araieval/ArGuard-2026-tasks
165
+ - **Submission — Subtask A1:** https://www.codabench.org/competitions/16909/
166
+ - **Submission — Subtask A2:** https://www.codabench.org/competitions/16910/
167
  - **Contact organisers:** arguard2026-organizers@googlegroups.com
168
 
169
  ## Annotation
 
196
 
197
  ## Citation
198
 
199
+ A citation will be provided when the shared-task overview paper is
200
+ released. Until then, please cite this repository URL.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
201
 
202
  ## Contact
203
 
204
  - **Email:** arguard2026-organizers@googlegroups.com
205
+ - **Website:** https://araieval.github.io/ArGuard2026/
206
  - **GitHub:** https://github.com/araieval/ArGuard-2026-tasks