Datasets:

Formats:
parquet
Languages:
English
Size:
< 1K
ArXiv:
License:
Samoed commited on
Commit
00a4288
·
verified ·
1 Parent(s): 3a5cdb8

Add dataset card

Browse files
Files changed (1) hide show
  1. README.md +310 -0
README.md CHANGED
@@ -1,4 +1,15 @@
1
  ---
 
 
 
 
 
 
 
 
 
 
 
2
  dataset_info:
3
  - config_name: corpus
4
  features:
@@ -55,4 +66,303 @@ configs:
55
  data_files:
56
  - split: test
57
  path: queries/test-*
 
 
 
 
 
58
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ annotations_creators:
3
+ - human-annotated
4
+ language:
5
+ - eng
6
+ license: unknown
7
+ multilinguality: monolingual
8
+ source_datasets:
9
+ - mteb/AVE-Dataset
10
+ task_categories:
11
+ - video-classification
12
+ task_ids: []
13
  dataset_info:
14
  - config_name: corpus
15
  features:
 
66
  data_files:
67
  - split: test
68
  path: queries/test-*
69
+ tags:
70
+ - mteb
71
+ - video
72
+ - audio
73
+ - text
74
  ---
75
+ <!-- adapted from https://github.com/huggingface/huggingface_hub/blob/v0.30.2/src/huggingface_hub/templates/datasetcard_template.md -->
76
+
77
+ <div align="center" style="padding: 40px 20px; background-color: white; border-radius: 12px; box-shadow: 0 2px 10px rgba(0, 0, 0, 0.05); max-width: 600px; margin: 0 auto;">
78
+ <h1 style="font-size: 3.5rem; color: #1a1a1a; margin: 0 0 20px 0; letter-spacing: 2px; font-weight: 700;">AVEDatasetZeroShot</h1>
79
+ <div style="font-size: 1.5rem; color: #4a4a4a; margin-bottom: 5px; font-weight: 300;">An <a href="https://github.com/embeddings-benchmark/mteb" style="color: #2c5282; font-weight: 600; text-decoration: none;" onmouseover="this.style.textDecoration='underline'" onmouseout="this.style.textDecoration='none'">MTEB</a> dataset</div>
80
+ <div style="font-size: 0.9rem; color: #2c5282; margin-top: 10px;">Massive Text Embedding Benchmark</div>
81
+ </div>
82
+
83
+ Audio-Visual Event (AVE) classification dataset containing 28 event categories such as church bell, truck, dog barking, and other everyday sounds sourced from YouTube videos. The goal is to predict the sound event category from synchronized video and audio. This variant uses both video and audio modalities. Used the official train and test splits across 28 audio-visual event classes (~3,312 train / ~402 test).
84
+
85
+ | | |
86
+ |---------------|---------------------------------------------|
87
+ | Task category | VideoZeroshotClassification (video+audio-to-text) |
88
+ | Domains | AudioScene, Web |
89
+ | Reference | [The European Conference on Computer Vision (ECCV)](https://openaccess.thecvf.com/content_ECCV_2018/papers/Yapeng_Tian_Audio-Visual_Event_Localization_ECCV_2018_paper.pdf) |
90
+
91
+ Source datasets:
92
+ - [mteb/AVE-Dataset](https://huggingface.co/datasets/mteb/AVE-Dataset)
93
+
94
+
95
+ ## How to evaluate on this task
96
+
97
+ You can evaluate an embedding model on this dataset using the following code:
98
+
99
+ ```python
100
+ import mteb
101
+
102
+ task = mteb.get_task("AVEDatasetZeroShot")
103
+ model = mteb.get_model(YOUR_MODEL)
104
+ mteb.evaluate(model, task)
105
+ ```
106
+
107
+ <!-- Datasets want link to arxiv in readme to autolink dataset with paper -->
108
+ To learn more about how to run models on `mteb` task check out the [GitHub repository](https://github.com/embeddings-benchmark/mteb).
109
+
110
+ ## Citation
111
+
112
+ If you use this dataset, please cite the dataset as well as [mteb](https://github.com/embeddings-benchmark/mteb), as this dataset likely includes additional processing as a part of the [MMTEB Contribution](https://github.com/embeddings-benchmark/mteb/tree/main/docs/mmteb).
113
+
114
+ ```bibtex
115
+
116
+ @inproceedings{Tian_2018_ECCV,
117
+ author = {Tian, Yapeng and Shi, Jing and Li, Bochen and Duan, Zhiyao and Xu, Chenliang},
118
+ booktitle = {The European Conference on Computer Vision (ECCV)},
119
+ month = {September},
120
+ title = {Audio-Visual Event Localization in Unconstrained Videos},
121
+ year = {2018},
122
+ }
123
+
124
+
125
+ @article{enevoldsen2025mmtebmassivemultilingualtext,
126
+ title={MMTEB: Massive Multilingual Text Embedding Benchmark},
127
+ author={Kenneth Enevoldsen and Isaac Chung and Imene Kerboua and Márton Kardos and Ashwin Mathur and David Stap and Jay Gala and Wissam Siblini and Dominik Krzemiński and Genta Indra Winata and Saba Sturua and Saiteja Utpala and Mathieu Ciancone and Marion Schaeffer and Gabriel Sequeira and Diganta Misra and Shreeya Dhakal and Jonathan Rystrøm and Roman Solomatin and Ömer Çağatan and Akash Kundu and Martin Bernstorff and Shitao Xiao and Akshita Sukhlecha and Bhavish Pahwa and Rafał Poświata and Kranthi Kiran GV and Shawon Ashraf and Daniel Auras and Björn Plüster and Jan Philipp Harries and Loïc Magne and Isabelle Mohr and Mariya Hendriksen and Dawei Zhu and Hippolyte Gisserot-Boukhlef and Tom Aarsen and Jan Kostkan and Konrad Wojtasik and Taemin Lee and Marek Šuppa and Crystina Zhang and Roberta Rocca and Mohammed Hamdy and Andrianos Michail and John Yang and Manuel Faysse and Aleksei Vatolin and Nandan Thakur and Manan Dey and Dipam Vasani and Pranjal Chitale and Simone Tedeschi and Nguyen Tai and Artem Snegirev and Michael Günther and Mengzhou Xia and Weijia Shi and Xing Han Lù and Jordan Clive and Gayatri Krishnakumar and Anna Maksimova and Silvan Wehrli and Maria Tikhonova and Henil Panchal and Aleksandr Abramov and Malte Ostendorff and Zheng Liu and Simon Clematide and Lester James Miranda and Alena Fenogenova and Guangyu Song and Ruqiya Bin Safi and Wen-Ding Li and Alessia Borghini and Federico Cassano and Hongjin Su and Jimmy Lin and Howard Yen and Lasse Hansen and Sara Hooker and Chenghao Xiao and Vaibhav Adlakha and Orion Weller and Siva Reddy and Niklas Muennighoff},
128
+ publisher = {arXiv},
129
+ journal={arXiv preprint arXiv:2502.13595},
130
+ year={2025},
131
+ url={https://arxiv.org/abs/2502.13595},
132
+ doi = {10.48550/arXiv.2502.13595},
133
+ }
134
+
135
+ @article{muennighoff2022mteb,
136
+ author = {Muennighoff, Niklas and Tazi, Nouamane and Magne, Loïc and Reimers, Nils},
137
+ title = {MTEB: Massive Text Embedding Benchmark},
138
+ publisher = {arXiv},
139
+ journal={arXiv preprint arXiv:2210.07316},
140
+ year = {2022}
141
+ url = {https://arxiv.org/abs/2210.07316},
142
+ doi = {10.48550/ARXIV.2210.07316},
143
+ }
144
+ ```
145
+
146
+ # Dataset Statistics
147
+ <details>
148
+ <summary> Dataset Statistics</summary>
149
+
150
+ The following code contains the descriptive statistics from the task. These can also be obtained using:
151
+
152
+ ```python
153
+ import mteb
154
+
155
+ task = mteb.get_task("AVEDatasetZeroShot")
156
+
157
+ desc_stats = task.metadata.descriptive_stats
158
+ ```
159
+
160
+ ```json
161
+ {
162
+ "test": {
163
+ "num_samples": 402,
164
+ "text_statistics": null,
165
+ "image_statistics": null,
166
+ "audio_statistics": {
167
+ "total_duration_seconds": 4025.644374999997,
168
+ "min_duration_seconds": 9.9845625,
169
+ "average_duration_seconds": 10.01404073383084,
170
+ "max_duration_seconds": 10.05425,
171
+ "unique_audios": 402,
172
+ "average_sampling_rate": 16000.0,
173
+ "sampling_rates": {
174
+ "16000": 402
175
+ }
176
+ },
177
+ "video_statistics": {
178
+ "total_duration_seconds": 4026.542646000018,
179
+ "total_frames": 115005,
180
+ "min_width": 192,
181
+ "average_width": 769.9601990049752,
182
+ "max_width": 1280,
183
+ "min_height": 144,
184
+ "average_height": 484.2587064676617,
185
+ "max_height": 720,
186
+ "min_duration_seconds": 10.0,
187
+ "average_duration_seconds": 10.016275238806015,
188
+ "max_duration_seconds": 10.061281,
189
+ "unique_videos": 402,
190
+ "average_fps": 28.579601990049753,
191
+ "fps": {
192
+ "30": 289,
193
+ "25": 87,
194
+ "24": 18,
195
+ "27": 1,
196
+ "29": 4,
197
+ "21": 1,
198
+ "28": 1,
199
+ "20": 1
200
+ },
201
+ "min_resolution": [
202
+ 192,
203
+ 144
204
+ ],
205
+ "average_resolution": [
206
+ 769.9601990049752,
207
+ 484.2587064676617
208
+ ],
209
+ "max_resolution": [
210
+ 1280,
211
+ 720
212
+ ],
213
+ "resolutions": {
214
+ "640x360": 25,
215
+ "1280x720": 143,
216
+ "480x360": 112,
217
+ "640x352": 3,
218
+ "960x720": 9,
219
+ "490x360": 1,
220
+ "480x270": 3,
221
+ "320x240": 30,
222
+ "512x288": 2,
223
+ "480x352": 1,
224
+ "540x360": 11,
225
+ "492x360": 3,
226
+ "406x720": 3,
227
+ "450x360": 6,
228
+ "640x354": 1,
229
+ "484x360": 1,
230
+ "352x288": 2,
231
+ "322x240": 1,
232
+ "360x288": 1,
233
+ "384x288": 5,
234
+ "192x144": 2,
235
+ "480x272": 2,
236
+ "352x240": 2,
237
+ "1080x720": 1,
238
+ "568x320": 3,
239
+ "640x290": 1,
240
+ "270x360": 2,
241
+ "638x360": 1,
242
+ "636x360": 1,
243
+ "426x240": 2,
244
+ "720x720": 1,
245
+ "422x360": 1,
246
+ "480x268": 1,
247
+ "480x318": 1,
248
+ "1170x720": 1,
249
+ "480x266": 1,
250
+ "202x360": 2,
251
+ "400x328": 1,
252
+ "480x320": 4,
253
+ "204x360": 3,
254
+ "198x360": 1,
255
+ "312x234": 1,
256
+ "398x238": 1,
257
+ "352x264": 1,
258
+ "480x356": 1,
259
+ "400x226": 1
260
+ }
261
+ },
262
+ "label_statistics": {
263
+ "min_labels_per_text": 1,
264
+ "average_label_per_text": 1.0,
265
+ "max_labels_per_text": 1,
266
+ "unique_labels": 28,
267
+ "labels": {
268
+ "8": {
269
+ "count": 18
270
+ },
271
+ "17": {
272
+ "count": 17
273
+ },
274
+ "4": {
275
+ "count": 17
276
+ },
277
+ "11": {
278
+ "count": 18
279
+ },
280
+ "20": {
281
+ "count": 18
282
+ },
283
+ "10": {
284
+ "count": 17
285
+ },
286
+ "15": {
287
+ "count": 17
288
+ },
289
+ "27": {
290
+ "count": 18
291
+ },
292
+ "12": {
293
+ "count": 17
294
+ },
295
+ "26": {
296
+ "count": 16
297
+ },
298
+ "13": {
299
+ "count": 18
300
+ },
301
+ "25": {
302
+ "count": 13
303
+ },
304
+ "22": {
305
+ "count": 10
306
+ },
307
+ "19": {
308
+ "count": 9
309
+ },
310
+ "1": {
311
+ "count": 18
312
+ },
313
+ "24": {
314
+ "count": 17
315
+ },
316
+ "9": {
317
+ "count": 16
318
+ },
319
+ "3": {
320
+ "count": 18
321
+ },
322
+ "14": {
323
+ "count": 10
324
+ },
325
+ "2": {
326
+ "count": 8
327
+ },
328
+ "5": {
329
+ "count": 6
330
+ },
331
+ "7": {
332
+ "count": 16
333
+ },
334
+ "6": {
335
+ "count": 6
336
+ },
337
+ "16": {
338
+ "count": 6
339
+ },
340
+ "23": {
341
+ "count": 18
342
+ },
343
+ "21": {
344
+ "count": 10
345
+ },
346
+ "0": {
347
+ "count": 15
348
+ },
349
+ "18": {
350
+ "count": 15
351
+ }
352
+ }
353
+ },
354
+ "candidates_labels_text_statistics": {
355
+ "total_text_length": 623,
356
+ "min_text_length": 14,
357
+ "average_text_length": 22.25,
358
+ "max_text_length": 40,
359
+ "unique_texts": 28
360
+ }
361
+ }
362
+ }
363
+ ```
364
+
365
+ </details>
366
+
367
+ ---
368
+ *This dataset card was automatically generated using [MTEB](https://github.com/embeddings-benchmark/mteb)*