MMTEB: Massive Multilingual Text Embedding Benchmark
Paper • 2502.13595 • Published • 50
Datasets:
id string | text string |
|---|---|
0 | a video of ApplyEyeMakeup |
1 | a video of ApplyLipstick |
2 | a video of Archery |
3 | a video of BabyCrawling |
4 | a video of BalanceBeam |
5 | a video of BandMarching |
6 | a video of BasketballDunk |
7 | a video of BlowDryHair |
8 | a video of BlowingCandles |
9 | a video of BodyWeightSquats |
10 | a video of Bowling |
11 | a video of BoxingPunchingBag |
12 | a video of BoxingSpeedBag |
13 | a video of BrushingTeeth |
14 | a video of CliffDiving |
15 | a video of CricketBowling |
16 | a video of CricketShot |
17 | a video of CuttingInKitchen |
18 | a video of FieldHockeyPenalty |
19 | a video of FloorGymnastics |
20 | a video of FrisbeeCatch |
21 | a video of FrontCrawl |
22 | a video of Haircut |
23 | a video of HammerThrow |
24 | a video of Hammering |
25 | a video of HandstandPushups |
26 | a video of HandstandWalking |
27 | a video of HeadMassage |
28 | a video of IceDancing |
29 | a video of Knitting |
30 | a video of LongJump |
31 | a video of MoppingFloor |
32 | a video of ParallelBars |
33 | a video of PlayingCello |
34 | a video of PlayingDaf |
35 | a video of PlayingDhol |
36 | a video of PlayingFlute |
37 | a video of PlayingSitar |
38 | a video of Rafting |
39 | a video of ShavingBeard |
40 | a video of Shotput |
41 | a video of SkyDiving |
42 | a video of SoccerPenalty |
43 | a video of StillRings |
44 | a video of SumoWrestling |
45 | a video of Surfing |
46 | a video of TableTennisShot |
47 | a video of Typing |
48 | a video of UnevenBars |
49 | a video of WallPushups |
50 | a video of WritingOnBoard |
Classifying video clips with audio into 51 human action categories from the UCF101 dataset. Used official split 1 restricted to a 51-class audio-bearing subset (~4,890 train / ~1,944 test).
| Task category | VideoZeroshotClassification (video+audio-to-text) |
| Domains | Activity, Sport, Web |
| Reference | UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild |
Source datasets:
How to evaluate on this task
You can evaluate an embedding model on this dataset using the following code:
import mteb
task = mteb.get_task("UCF101VideoZeroShotClassification")
model = mteb.get_model(YOUR_MODEL)
mteb.evaluate(model, task)
To learn more about how to run models on mteb task check out the GitHub repository.
Citation
If you use this dataset, please cite the dataset as well as mteb, as this dataset likely includes additional processing as a part of the MMTEB Contribution.
@misc{soomro2012ucf101dataset101human,
archiveprefix = {arXiv},
author = {Khurram Soomro and Amir Roshan Zamir and Mubarak Shah},
eprint = {1212.0402},
primaryclass = {cs.CV},
title = {UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild},
url = {https://arxiv.org/abs/1212.0402},
year = {2012},
}
@article{enevoldsen2025mmtebmassivemultilingualtext,
title={MMTEB: Massive Multilingual Text Embedding Benchmark},
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},
publisher = {arXiv},
journal={arXiv preprint arXiv:2502.13595},
year={2025},
url={https://arxiv.org/abs/2502.13595},
doi = {10.48550/arXiv.2502.13595},
}
@article{muennighoff2022mteb,
author = {Muennighoff, Niklas and Tazi, Nouamane and Magne, Loïc and Reimers, Nils},
title = {MTEB: Massive Text Embedding Benchmark},
publisher = {arXiv},
journal={arXiv preprint arXiv:2210.07316},
year = {2022}
url = {https://arxiv.org/abs/2210.07316},
doi = {10.48550/ARXIV.2210.07316},
}
Dataset Statistics
Dataset Statistics
The following code contains the descriptive statistics from the task. These can also be obtained using:
import mteb
task = mteb.get_task("UCF101VideoZeroShotClassification")
desc_stats = task.metadata.descriptive_stats
{
"test": {
"num_samples": 1944,
"text_statistics": null,
"image_statistics": null,
"audio_statistics": null,
"video_statistics": {
"total_duration_seconds": 13304.320000000025,
"total_frames": 332608,
"min_width": 320,
"average_width": 320.0,
"max_width": 320,
"min_height": 240,
"average_height": 240.0,
"max_height": 240,
"min_duration_seconds": 1.2,
"average_duration_seconds": 6.843786008230466,
"max_duration_seconds": 31.72,
"unique_videos": 1942,
"average_fps": 25.0,
"fps": {
"25": 1944
},
"min_resolution": [
320,
240
],
"average_resolution": [
320.0,
240.0
],
"max_resolution": [
320,
240
],
"resolutions": {
"320x240": 1944
}
},
"label_statistics": {
"min_labels_per_text": 1,
"average_label_per_text": 1.0,
"max_labels_per_text": 1,
"unique_labels": 51,
"labels": {
"0": {
"count": 44
},
"1": {
"count": 32
},
"2": {
"count": 41
},
"3": {
"count": 35
},
"4": {
"count": 31
},
"5": {
"count": 43
},
"6": {
"count": 37
},
"7": {
"count": 38
},
"8": {
"count": 33
},
"9": {
"count": 30
},
"10": {
"count": 43
},
"11": {
"count": 49
},
"12": {
"count": 37
},
"13": {
"count": 36
},
"14": {
"count": 39
},
"15": {
"count": 36
},
"16": {
"count": 49
},
"17": {
"count": 33
},
"18": {
"count": 40
},
"19": {
"count": 36
},
"20": {
"count": 37
},
"21": {
"count": 37
},
"22": {
"count": 33
},
"24": {
"count": 33
},
"23": {
"count": 45
},
"25": {
"count": 28
},
"26": {
"count": 34
},
"27": {
"count": 41
},
"28": {
"count": 46
},
"29": {
"count": 34
},
"30": {
"count": 39
},
"31": {
"count": 34
},
"32": {
"count": 37
},
"33": {
"count": 44
},
"34": {
"count": 41
},
"35": {
"count": 49
},
"36": {
"count": 48
},
"37": {
"count": 44
},
"38": {
"count": 28
},
"39": {
"count": 43
},
"40": {
"count": 46
},
"41": {
"count": 31
},
"42": {
"count": 41
},
"43": {
"count": 32
},
"44": {
"count": 34
},
"45": {
"count": 33
},
"46": {
"count": 39
},
"47": {
"count": 43
},
"48": {
"count": 28
},
"49": {
"count": 35
},
"50": {
"count": 45
}
}
},
"candidates_labels_text_statistics": {
"total_text_length": 1154,
"min_text_length": 17,
"average_text_length": 22.627450980392158,
"max_text_length": 29,
"unique_texts": 51
}
}
}
This dataset card was automatically generated using MTEB
- Downloads last month
- 47