Skip to content

dataset: Add EuroPIRQRetrieval dataset #3924

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Samoed merged 2 commits into from
Jan 14, 2026
+161 −0
Merged

dataset: Add EuroPIRQRetrieval dataset #3924

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
41f2fd8
dataset: Add EuroPIRQRetrieval dataset
eherra Jan 12, 2026
e013eb1
Removed unnecessary load dataset functions
eherra Jan 13, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
116 changes: 116 additions & 0 deletions mteb/descriptive_stats/Retrieval/EuroPIRQRetrieval.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,116 @@
{
"test": {
"num_samples": 30300,
"number_of_characters": 17320243,
"documents_text_statistics": {
"total_text_length": 17276572,
"min_text_length": 316,
"average_text_length": 575.8857333333333,
"max_text_length": 1008,
"unique_texts": 28361
},
"documents_image_statistics": null,
"queries_text_statistics": {
"total_text_length": 43671,
"min_text_length": 67,
"average_text_length": 145.57,
"max_text_length": 345,
"unique_texts": 300
},
"queries_image_statistics": null,
"relevant_docs_statistics": {
"num_relevant_docs": 300,
"min_relevant_docs_per_query": 1,
"average_relevant_docs_per_query": 1.0,
"max_relevant_docs_per_query": 1,
"unique_relevant_docs": 300
},
"top_ranked_statistics": null,
"hf_subset_descriptive_stats": {
"en": {
"num_samples": 10100,
"number_of_characters": 5517678,
"documents_text_statistics": {
"total_text_length": 5503635,
"min_text_length": 316,
"average_text_length": 550.3635,
"max_text_length": 726,
"unique_texts": 9422
},
"documents_image_statistics": null,
"queries_text_statistics": {
"total_text_length": 14043,
"min_text_length": 68,
"average_text_length": 140.43,
"max_text_length": 305,
"unique_texts": 100
},
"queries_image_statistics": null,
"relevant_docs_statistics": {
"num_relevant_docs": 100,
"min_relevant_docs_per_query": 1,
"average_relevant_docs_per_query": 1.0,
"max_relevant_docs_per_query": 1,
"unique_relevant_docs": 100
},
"top_ranked_statistics": null
},
"fi": {
"num_samples": 10100,
"number_of_characters": 5953462,
"documents_text_statistics": {
"total_text_length": 5938809,
"min_text_length": 326,
"average_text_length": 593.8809,
"max_text_length": 1008,
"unique_texts": 9422
},
"documents_image_statistics": null,
"queries_text_statistics": {
"total_text_length": 14653,
"min_text_length": 67,
"average_text_length": 146.53,
"max_text_length": 345,
"unique_texts": 100
},
"queries_image_statistics": null,
"relevant_docs_statistics": {
"num_relevant_docs": 100,
"min_relevant_docs_per_query": 1,
"average_relevant_docs_per_query": 1.0,
"max_relevant_docs_per_query": 1,
"unique_relevant_docs": 100
},
"top_ranked_statistics": null
},
"pt": {
"num_samples": 10100,
"number_of_characters": 5849103,
"documents_text_statistics": {
"total_text_length": 5834128,
"min_text_length": 325,
"average_text_length": 583.4128,
"max_text_length": 774,
"unique_texts": 9517
},
"documents_image_statistics": null,
"queries_text_statistics": {
"total_text_length": 14975,
"min_text_length": 69,
"average_text_length": 149.75,
"max_text_length": 320,
"unique_texts": 100
},
"queries_image_statistics": null,
"relevant_docs_statistics": {
"num_relevant_docs": 100,
"min_relevant_docs_per_query": 1,
"average_relevant_docs_per_query": 1.0,
"max_relevant_docs_per_query": 1,
"unique_relevant_docs": 100
},
"top_ranked_statistics": null
}
}
}
}
2 changes: 2 additions & 0 deletions mteb/tasks/retrieval/multilingual/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
CrossLingualSemanticDiscriminationWMT21,
)
from .cur_ev1_retrieval import CUREv1Retrieval
from .euro_pirq_retrieval import EuroPIRQRetrieval
from .indic_qa_retrieval import IndicQARetrieval
from .jina_vdr_bench_retrieval import (
JinaVDRAirbnbSyntheticRetrieval,
Expand Down Expand Up @@ -107,6 +108,7 @@
"CUREv1Retrieval",
"CrossLingualSemanticDiscriminationWMT19",
"CrossLingualSemanticDiscriminationWMT21",
"EuroPIRQRetrieval",
"IndicQARetrieval",
"JinaVDRAirbnbSyntheticRetrieval",
"JinaVDRArabicChartQARetrieval",
Expand Down
43 changes: 43 additions & 0 deletions mteb/tasks/retrieval/multilingual/euro_pirq_retrieval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,43 @@
from mteb.abstasks.retrieval import AbsTaskRetrieval
from mteb.abstasks.task_metadata import TaskMetadata

_LANGUAGES = {
"en": ["eng-Latn"],
"fi": ["fin-Latn"],
"pt": ["por-Latn"],
}


class EuroPIRQRetrieval(AbsTaskRetrieval):
metadata = TaskMetadata(
name="EuroPIRQRetrieval",
description="The EuroPIRQ retrieval dataset is a multilingual collection designed for evaluating retrieval and cross-lingual retrieval tasks. Dataset contains 10,000 parallel passages & 100 parallel queries (synthetic) in three languages: English, Portuguese, and Finnish, constructed from the European Union's DGT-Acquis corpus.",
reference="https://huggingface.co/datasets/eherra/EuroPIRQ-retrieval",
dataset={
"path": "eherra/EuroPIRQ-retrieval",
"revision": "59225ed25fbcea2185e1acbc8c3c80f1a8cd8341",
},
type="Retrieval",
category="t2t",
modalities=["text"],
eval_splits=["test"],
eval_langs=_LANGUAGES,
main_score="ndcg_at_10",
date=("2025-12-01", "2025-12-31"),
domains=["Legal"],
task_subtypes=[],
license="not specified",
annotations_creators="LM-generated and reviewed",
dialect=[],
sample_creation="found",
is_public=True,
bibtex_citation=r"""
@misc{eherra_2025_europirq,
author = { {Elias Herranen} },
publisher = { Hugging Face },
title = { EuroPIRQ: European Parallel Information Retrieval Queries },
url = { https://huggingface.co/datasets/eherra/EuroPIRQ-retrieval },
year = {2025},
}
""",
)