Datasets:

Modalities:
Text
Formats:
parquet
ArXiv:
Libraries:
Datasets
pandas
License:
ESCIReranking / README.md
Samoed's picture
Add dataset card
e459d89 verified
metadata
annotations_creators:
  - derived
language:
  - eng
  - jpn
  - spa
license: apache-2.0
multilinguality: multilingual
task_categories:
  - text-ranking
task_ids: []
dataset_info:
  - config_name: es-corpus
    features:
      - name: _id
        dtype: string
      - name: text
        dtype: string
      - name: title
        dtype: string
    splits:
      - name: train
        num_bytes: 88812088
        num_examples: 83592
      - name: test
        num_bytes: 34083231
        num_examples: 31934
    download_size: 55938472
    dataset_size: 122895319
  - config_name: es-qrels
    features:
      - name: query-id
        dtype: string
      - name: corpus-id
        dtype: string
      - name: score
        dtype: int64
    splits:
      - name: train
        num_bytes: 5111371
        num_examples: 83592
      - name: test
        num_bytes: 1864847
        num_examples: 31934
    download_size: 850278
    dataset_size: 6976218
  - config_name: es-queries
    features:
      - name: _id
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: train
        num_bytes: 216721
        num_examples: 4952
      - name: test
        num_bytes: 79213
        num_examples: 1851
    download_size: 173667
    dataset_size: 295934
  - config_name: es-top_ranked
    features:
      - name: query-id
        dtype: string
      - name: corpus-ids
        sequence: string
    splits:
      - name: train
        num_bytes: 2990077
        num_examples: 4952
      - name: test
        num_bytes: 1094946
        num_examples: 1851
    download_size: 820067
    dataset_size: 4085023
  - config_name: jp-corpus
    features:
      - name: _id
        dtype: string
      - name: text
        dtype: string
      - name: title
        dtype: string
    splits:
      - name: train
        num_bytes: 68820021
        num_examples: 74668
      - name: test
        num_bytes: 26851604
        num_examples: 29096
    download_size: 41580610
    dataset_size: 95671625
  - config_name: jp-qrels
    features:
      - name: query-id
        dtype: string
      - name: corpus-id
        dtype: string
      - name: score
        dtype: int64
    splits:
      - name: train
        num_bytes: 4566293
        num_examples: 74668
      - name: test
        num_bytes: 1699622
        num_examples: 29096
    download_size: 771757
    dataset_size: 6265915
  - config_name: jp-queries
    features:
      - name: _id
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: train
        num_bytes: 241851
        num_examples: 4867
      - name: test
        num_bytes: 88153
        num_examples: 1850
    download_size: 193271
    dataset_size: 330004
  - config_name: jp-top_ranked
    features:
      - name: query-id
        dtype: string
      - name: corpus-ids
        sequence: string
    splits:
      - name: train
        num_bytes: 2682685
        num_examples: 4867
      - name: test
        num_bytes: 1001897
        num_examples: 1850
    download_size: 738007
    dataset_size: 3684582
  - config_name: us-corpus
    features:
      - name: _id
        dtype: string
      - name: text
        dtype: string
      - name: title
        dtype: string
    splits:
      - name: train
        num_bytes: 227494355
        num_examples: 250518
      - name: test
        num_bytes: 79000358
        num_examples: 87202
    download_size: 146805059
    dataset_size: 306494713
  - config_name: us-qrels
    features:
      - name: query-id
        dtype: string
      - name: corpus-id
        dtype: string
      - name: score
        dtype: int64
    splits:
      - name: train
        num_bytes: 15635975
        num_examples: 250518
      - name: test
        num_bytes: 5170331
        num_examples: 87202
    download_size: 2569662
    dataset_size: 20806306
  - config_name: us-queries
    features:
      - name: _id
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: train
        num_bytes: 890520
        num_examples: 19517
      - name: test
        num_bytes: 297220
        num_examples: 6694
    download_size: 689203
    dataset_size: 1187740
  - config_name: us-top_ranked
    features:
      - name: query-id
        dtype: string
      - name: corpus-ids
        sequence: string
    splits:
      - name: train
        num_bytes: 9244508
        num_examples: 19517
      - name: test
        num_bytes: 3068148
        num_examples: 6694
    download_size: 2440891
    dataset_size: 12312656
configs:
  - config_name: es-corpus
    data_files:
      - split: train
        path: es-corpus/train-*
      - split: test
        path: es-corpus/test-*
  - config_name: es-qrels
    data_files:
      - split: train
        path: es-qrels/train-*
      - split: test
        path: es-qrels/test-*
  - config_name: es-queries
    data_files:
      - split: train
        path: es-queries/train-*
      - split: test
        path: es-queries/test-*
  - config_name: es-top_ranked
    data_files:
      - split: train
        path: es-top_ranked/train-*
      - split: test
        path: es-top_ranked/test-*
  - config_name: jp-corpus
    data_files:
      - split: train
        path: jp-corpus/train-*
      - split: test
        path: jp-corpus/test-*
  - config_name: jp-qrels
    data_files:
      - split: train
        path: jp-qrels/train-*
      - split: test
        path: jp-qrels/test-*
  - config_name: jp-queries
    data_files:
      - split: train
        path: jp-queries/train-*
      - split: test
        path: jp-queries/test-*
  - config_name: jp-top_ranked
    data_files:
      - split: train
        path: jp-top_ranked/train-*
      - split: test
        path: jp-top_ranked/test-*
  - config_name: us-corpus
    data_files:
      - split: train
        path: us-corpus/train-*
      - split: test
        path: us-corpus/test-*
  - config_name: us-qrels
    data_files:
      - split: train
        path: us-qrels/train-*
      - split: test
        path: us-qrels/test-*
  - config_name: us-queries
    data_files:
      - split: train
        path: us-queries/train-*
      - split: test
        path: us-queries/test-*
  - config_name: us-top_ranked
    data_files:
      - split: train
        path: us-top_ranked/train-*
      - split: test
        path: us-top_ranked/test-*
tags:
  - mteb
  - text

ESCIReranking

An MTEB dataset
Massive Text Embedding Benchmark
Task category t2t
Domains Written
Reference https://github.com/amazon-science/esci-data/

How to evaluate on this task

You can evaluate an embedding model on this dataset using the following code:

import mteb

task = mteb.get_tasks(["ESCIReranking"])
evaluator = mteb.MTEB(task)

model = mteb.get_model(YOUR_MODEL)
evaluator.run(model)

To learn more about how to run models on mteb task check out the GitHub repitory.

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.

@article{reddy2022shopping,
  archiveprefix = {arXiv},
  author = {Chandan K. Reddy and Lluís Màrquez and Fran Valero and Nikhil Rao and Hugo Zaragoza and Sambaran Bandyopadhyay and Arnab Biswas and Anlu Xing and Karthik Subbian},
  eprint = {2206.06588},
  title = {Shopping Queries Dataset: A Large-Scale {ESCI} Benchmark for Improving Product Search},
  year = {2022},
}

@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, Lo{\"\i}c 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("ESCIReranking")

desc_stats = task.metadata.descriptive_stats
{
    "test": {
        "num_samples": 158627,
        "number_of_characters": 117894609,
        "num_documents": 148232,
        "min_document_length": 1,
        "average_document_length": 793.9222570025365,
        "max_document_length": 8640,
        "unique_documents": 148232,
        "num_queries": 10395,
        "min_query_length": 1,
        "average_query_length": 20.194805194805195,
        "max_query_length": 143,
        "unique_queries": 10395,
        "none_queries": 0,
        "num_relevant_docs": 148232,
        "min_relevant_docs_per_query": 2,
        "average_relevant_docs_per_query": 10.277825877825878,
        "max_relevant_docs_per_query": 74,
        "unique_relevant_docs": 148232,
        "num_instructions": null,
        "min_instruction_length": null,
        "average_instruction_length": null,
        "max_instruction_length": null,
        "unique_instructions": null,
        "num_top_ranked": 10395,
        "min_top_ranked_per_query": 2,
        "average_top_ranked_per_query": 14.25993265993266,
        "max_top_ranked_per_query": 74,
        "hf_subset_descriptive_stats": {
            "us": {
                "num_samples": 93896,
                "number_of_characters": 74993786,
                "num_documents": 87202,
                "min_document_length": 1,
                "average_document_length": 858.2693745556295,
                "max_document_length": 4608,
                "unique_documents": 87202,
                "num_queries": 6694,
                "min_query_length": 1,
                "average_query_length": 22.554526441589484,
                "max_query_length": 143,
                "unique_queries": 6694,
                "none_queries": 0,
                "num_relevant_docs": 87202,
                "min_relevant_docs_per_query": 2,
                "average_relevant_docs_per_query": 9.446519270988945,
                "max_relevant_docs_per_query": 74,
                "unique_relevant_docs": 87202,
                "num_instructions": null,
                "min_instruction_length": null,
                "average_instruction_length": null,
                "max_instruction_length": null,
                "unique_instructions": null,
                "num_top_ranked": 6694,
                "min_top_ranked_per_query": 2,
                "average_top_ranked_per_query": 13.026889752016732,
                "max_top_ranked_per_query": 74
            },
            "es": {
                "num_samples": 33785,
                "number_of_characters": 32170187,
                "num_documents": 31934,
                "min_document_length": 1,
                "average_document_length": 1006.1636500281832,
                "max_document_length": 8640,
                "unique_documents": 31934,
                "num_queries": 1851,
                "min_query_length": 3,
                "average_query_length": 21.262560777957862,
                "max_query_length": 59,
                "unique_queries": 1851,
                "none_queries": 0,
                "num_relevant_docs": 31934,
                "min_relevant_docs_per_query": 2,
                "average_relevant_docs_per_query": 12.038357644516477,
                "max_relevant_docs_per_query": 41,
                "unique_relevant_docs": 31934,
                "num_instructions": null,
                "min_instruction_length": null,
                "average_instruction_length": null,
                "max_instruction_length": null,
                "unique_instructions": null,
                "num_top_ranked": 1851,
                "min_top_ranked_per_query": 2,
                "average_top_ranked_per_query": 17.252296056185845,
                "max_top_ranked_per_query": 41
            },
            "jp": {
                "num_samples": 30946,
                "number_of_characters": 10730636,
                "num_documents": 29096,
                "min_document_length": 1,
                "average_document_length": 368.12785262579047,
                "max_document_length": 3940,
                "unique_documents": 29096,
                "num_queries": 1850,
                "min_query_length": 1,
                "average_query_length": 10.588108108108107,
                "max_query_length": 60,
                "unique_queries": 1850,
                "none_queries": 0,
                "num_relevant_docs": 29096,
                "min_relevant_docs_per_query": 2,
                "average_relevant_docs_per_query": 11.524324324324324,
                "max_relevant_docs_per_query": 50,
                "unique_relevant_docs": 29096,
                "num_instructions": null,
                "min_instruction_length": null,
                "average_instruction_length": null,
                "max_instruction_length": null,
                "unique_instructions": null,
                "num_top_ranked": 1850,
                "min_top_ranked_per_query": 2,
                "average_top_ranked_per_query": 15.727567567567567,
                "max_top_ranked_per_query": 50
            }
        }
    }
}

This dataset card was automatically generated using MTEB