metadata
annotations_creators:
- LM-generated and reviewed
language:
- eng
license: apache-2.0
multilinguality: monolingual
source_datasets:
- omkarthawakar/CoVR-R
task_categories:
- other
task_ids: []
dataset_info:
- config_name: corpus
features:
- name: video
dtype: video
- name: id
dtype: string
splits:
- name: test
num_bytes: 7540218516
num_examples: 4425
download_size: 5446810661
dataset_size: 7540218516
- config_name: qrels
features:
- name: query-id
dtype: string
- name: corpus-id
dtype: string
- name: score
dtype: int64
splits:
- name: test
num_bytes: 83558
num_examples: 2634
download_size: 52367
dataset_size: 83558
- config_name: queries
features:
- name: video
dtype: video
- name: source
dtype: string
- name: annotation_id
dtype: int64
- name: video_source
dtype: string
- name: video_target
dtype: string
- name: description_source
dtype: string
- name: description_target
dtype: string
- name: modification_text
dtype: string
- name: reasoned_target_video_description__main
dtype: string
- name: id_original
dtype: int64
- name: id
dtype: string
- name: text
dtype: string
splits:
- name: test
num_bytes: 5202852215
num_examples: 2634
download_size: 2809642976
dataset_size: 5202852215
configs:
- config_name: corpus
data_files:
- split: test
path: corpus/test-*
- config_name: qrels
data_files:
- split: test
path: qrels/test-*
- config_name: queries
data_files:
- split: test
path: queries/test-*
tags:
- mteb
- video
- text
CoVR-R is a reasoning-aware benchmark for composed video retrieval. Given a reference video and a textual modification, the goal is to retrieve the correct target video that reflects the requested change and its implied visual consequences.
| Task category | Any2AnyRetrieval (video+text-to-video) |
| Domains | Web, Activity |
| Reference | Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Findings |
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("CoVRRVT2VRetrieval")
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.
@inproceedings{thawakar2026covrr,
author = {Thawakar, Omkar and Demidov, Dmitry and Potlapalli, Vaishnav and Bogireddy, Sai Prasanna Teja Reddy and Gajjala, Viswanatha Reddy and Lasheen, Alaa Mostafa and Anwer, Rao Muhammad and Khan, Fahad Shahbaz},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Findings},
title = {CoVR-R: Reason-Aware Composed Video Retrieval},
year = {2026},
}
@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("CoVRRVT2VRetrieval")
desc_stats = task.metadata.descriptive_stats
{
"test": {
"num_samples": 7059,
"num_queries": 2634,
"num_documents": 4425,
"number_of_characters": 1348388,
"documents_text_statistics": null,
"documents_image_statistics": null,
"documents_audio_statistics": null,
"documents_video_statistics": {
"total_duration_seconds": 52576.90672017247,
"total_frames": 1397947,
"min_width": 240,
"average_width": 515.1430508474576,
"max_width": 600,
"min_height": 240,
"average_height": 297.4567231638418,
"max_height": 336,
"min_duration_seconds": 0.6666666666666666,
"average_duration_seconds": 11.881786829417507,
"max_duration_seconds": 104.0039,
"unique_videos": 4425,
"average_fps": 22.144858757062146,
"fps": {
"12": 1655,
"30": 1431,
"50": 55,
"25": 812,
"24": 327,
"60": 59,
"11": 12,
"9": 2,
"10": 30,
"4": 1,
"6": 7,
"7": 5,
"8": 29
},
"min_resolution": [
240,
240
],
"average_resolution": [
515.1430508474576,
297.4567231638418
],
"max_resolution": [
596,
336
],
"resolutions": {
"320x240": 384,
"432x240": 32,
"360x240": 132,
"427x240": 1010,
"596x336": 2403,
"600x316": 175,
"436x240": 5,
"592x336": 66,
"361x240": 8,
"293x240": 60,
"424x240": 66,
"446x336": 12,
"426x240": 21,
"504x336": 19,
"400x240": 5,
"417x240": 5,
"300x240": 2,
"483x240": 2,
"496x336": 4,
"600x322": 1,
"240x240": 2,
"420x240": 2,
"420x336": 4,
"458x336": 1,
"384x240": 2,
"429x240": 1,
"358x240": 1
}
},
"queries_text_statistics": {
"total_text_length": 1348388,
"min_text_length": 203,
"average_text_length": 511.916476841306,
"max_text_length": 902,
"unique_texts": 2634
},
"queries_image_statistics": null,
"queries_audio_statistics": null,
"queries_video_statistics": {
"total_duration_seconds": 28299.60347948147,
"total_frames": 730272,
"min_width": 240,
"average_width": 497.54365983295367,
"max_width": 600,
"min_height": 240,
"average_height": 289.0455580865604,
"max_height": 336,
"min_duration_seconds": 0.6666666666666666,
"average_duration_seconds": 10.7439648745184,
"max_duration_seconds": 88.42,
"unique_videos": 2621,
"average_fps": 20.675018982536066,
"fps": {
"25": 414,
"24": 167,
"30": 722,
"50": 32,
"60": 30,
"12": 1211,
"10": 22,
"11": 5,
"8": 21,
"7": 4,
"6": 4,
"9": 1,
"4": 1
},
"min_resolution": [
240,
240
],
"average_resolution": [
497.54365983295367,
289.0455580865604
],
"max_resolution": [
596,
336
],
"resolutions": {
"596x336": 1222,
"600x316": 92,
"592x336": 29,
"504x336": 11,
"496x336": 2,
"600x322": 1,
"446x336": 6,
"420x336": 1,
"458x336": 1,
"427x240": 739,
"320x240": 271,
"424x240": 51,
"360x240": 93,
"436x240": 5,
"426x240": 16,
"432x240": 23,
"417x240": 5,
"293x240": 48,
"420x240": 2,
"384x240": 2,
"300x240": 1,
"429x240": 1,
"361x240": 4,
"400x240": 4,
"240x240": 2,
"483x240": 1,
"358x240": 1
}
},
"relevant_docs_statistics": {
"num_relevant_docs": 2634,
"min_relevant_docs_per_query": 1,
"average_relevant_docs_per_query": 1.0,
"max_relevant_docs_per_query": 1,
"unique_relevant_docs": 2634
},
"top_ranked_statistics": null
}
}
This dataset card was automatically generated using MTEB