DocumentUnderstanding

• Number of tasks: 58

JinaVDRAirbnbSyntheticRetrieval

Retrieve rendered tables from Airbnb listings based on templated queries.

Dataset: jinaai/airbnb-synthetic-retrieval_beir • License: cc0-1.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	ara, deu, eng, fra, hin, ... (10)	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRArabicChartQARetrieval

Retrieve Arabic charts based on queries.

Dataset: jinaai/arabic_chartqa_ar_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	ara	Academic	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRArabicInfographicsVQARetrieval

Retrieve Arabic infographics based on queries.

Dataset: jinaai/arabic_infographicsvqa_ar_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	ara	Academic	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRArxivQARetrieval

Retrieve figures from scientific papers from arXiv based on LLM generated queries.

Dataset: jinaai/arxivqa_beir • License: cc-by-4.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRAutomobileCatelogRetrieval

Retrieve automobile marketing documents based on LLM generated queries.

Dataset: jinaai/automobile_catalogue_jp_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	jpn	Engineering, Web	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRBeveragesCatalogueRetrieval

Retrieve beverages marketing documents based on LLM generated queries.

Dataset: jinaai/beverages_catalogue_ru_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	rus	Web	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRCharXivOCRRetrieval

Retrieve charts from scientific papers based on human annotated queries.

Dataset: jinaai/CharXiv-en_beir • License: cc-by-sa-4.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRChartQARetrieval

Retrieve charts based on LLM generated queries.

Dataset: jinaai/ChartQA_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRDocQAAI

Retrieve AI documents based on LLM generated queries.

Dataset: jinaai/docqa_artificial_intelligence_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRDocQAEnergyRetrieval

Retrieve energy industry documents based on LLM generated queries.

Dataset: jinaai/docqa_energy_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRDocQAGovReportRetrieval

Retrieve government reports based on LLM generated queries.

Dataset: jinaai/docqa_gov_report_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Government	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRDocQAHealthcareIndustryRetrieval

Retrieve healthcare industry documents based on LLM generated queries.

Dataset: jinaai/docqa_healthcare_industry_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Medical	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRDocVQARetrieval

Retrieve industry documents based on human annotated queries.

Dataset: jinaai/docvqa_beir • License: cc-by-4.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRDonutVQAISynHMPRetrieval

Retrieve medical records based on templated queries.

Dataset: jinaai/donut_vqa_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Medical	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDREuropeanaDeNewsRetrieval

Retrieve German news articles based on LLM generated queries.

Dataset: jinaai/europeana-de-news_beir • License: cc0-1.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	deu	News	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDREuropeanaEsNewsRetrieval

Retrieve Spanish news articles based on LLM generated queries.

Dataset: jinaai/europeana-es-news_beir • License: cc0-1.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	spa	News	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDREuropeanaFrNewsRetrieval

Retrieve French news articles from Europeana based on LLM generated queries.

Dataset: jinaai/europeana-fr-news_beir • License: cc0-1.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	fra	News	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDREuropeanaItScansRetrieval

Retrieve Italian historical articles based on LLM generated queries.

Dataset: jinaai/europeana-it-scans_beir • License: cc0-1.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	ita	News	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDREuropeanaNlLegalRetrieval

Retrieve Dutch historical legal documents based on LLM generated queries.

Dataset: jinaai/europeana-nl-legal_beir • License: cc0-1.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	nld	Legal	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRGitHubReadmeRetrieval

Retrieve GitHub readme files based their description.

Dataset: jinaai/github-readme-retrieval-multilingual_beir • License: multiple • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	ara, ben, deu, eng, fra, ... (17)	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRHindiGovVQARetrieval

Retrieve Hindi government documents based on LLM generated queries.

Dataset: jinaai/hindi-gov-vqa_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	hin	Government	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRHungarianDocQARetrieval

Retrieve Hungarian documents in various formats based on human annotated queries.

Dataset: jinaai/hungarian_doc_qa_beir • License: cc-by-4.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	hun	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRInfovqaRetrieval

Retrieve infographics based on human annotated queries.

Dataset: jinaai/infovqa_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRJDocQARetrieval

Retrieve Japanese documents in various formats based on human annotated queries.

Dataset: jinaai/jdocqa_beir • License: cc-by-4.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	jpn	Web	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRJina2024YearlyBookRetrieval

Retrieve pages from the 2024 Jina yearbook based on human annotated questions.

Dataset: jinaai/jina_2024_yearly_book_beir • License: apache-2.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRMMTabRetrieval

Retrieve tables from the MMTab dataset based on queries.

Dataset: jinaai/MMTab_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRMPMQARetrieval

Retrieve product manuals based on human annotated queries.

Dataset: jinaai/mpmqa_small_beir • License: apache-2.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	human-annotated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRMedicalPrescriptionsRetrieval

Retrieve medical prescriptions based on templated queries.

Dataset: jinaai/medical-prescriptions_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Medical	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDROWIDChartsRetrieval

Retrieve charts from the OWID dataset based on accompanied text snippets.

Dataset: jinaai/owid_charts_en_beir • License: cc-by-4.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDROpenAINewsRetrieval

Retrieve news articles from the OpenAI news website based on human annotated queries.

Dataset: jinaai/openai-news_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	News, Web	human-annotated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRPlotQARetrieval

Retrieve plots from the PlotQA dataset based on LLM generated queries.

Dataset: jinaai/plotqa_beir • License: cc-by-4.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRRamensBenchmarkRetrieval

Retrieve ramen restaurant marketing documents based on LLM generated queries.

Dataset: jinaai/ramen_benchmark_jp_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	jpn	Web	LM-generated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRShanghaiMasterPlanRetrieval

Retrieve pages from the Shanghai Master Plan based on human annotated queries.

Dataset: jinaai/shanghai_master_plan_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	zho	Web	human-annotated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRShiftProjectRetrieval

Retrieve documents with graphs from the Shift Project based on LLM generated queries.

Dataset: jinaai/shiftproject_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRStanfordSlideRetrieval

Retrieve scientific and engineering slides based on human annotated queries.

Dataset: jinaai/stanford_slide_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	human-annotated	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRStudentEnrollmentSyntheticRetrieval

Retrieve student enrollment data based on templated queries.

Dataset: jinaai/student-enrollment_beir • License: cc0-1.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRTQARetrieval

Retrieve textbook pages (images and text) based on LLM generated queries from the text.

Dataset: jinaai/tqa_beir • License: cc-by-nc-3.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRTabFQuadRetrieval

Retrieve tables from industry documents based on LLM generated queries.

Dataset: jinaai/tabfquad_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRTableVQARetrieval

Retrieve scientific tables based on LLM generated queries.

Dataset: jinaai/table-vqa_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRTatQARetrieval

Retrieve financial reports based on human annotated queries.

Dataset: jinaai/tatqa_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRTweetStockSyntheticsRetrieval

Retrieve rendered tables of stock prices based on templated queries.

Dataset: jinaai/tweet-stock-synthetic-retrieval_beir • License: not specified • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	ara, deu, eng, fra, hin, ... (10)	Social	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRWikimediaCommonsDocumentsRetrieval

Retrieve historical documents from Wikimedia Commons based on their description.

Dataset: jinaai/wikimedia-commons-documents-ml_beir • License: multiple • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	ara, ben, deu, eng, fra, ... (20)	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

JinaVDRWikimediaCommonsMapsRetrieval

Retrieve maps from Wikimedia Commons based on their description.

Dataset: jinaai/wikimedia-commons-maps_beir • License: multiple • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Web	derived	found

Citation

@misc{günther2025jinaembeddingsv4universalembeddingsmultimodal,
  archiveprefix = {arXiv},
  author = {Michael Günther and Saba Sturua and Mohammad Kalim Akram and Isabelle Mohr and Andrei Ungureanu and Bo Wang and Sedigheh Eslami and Scott Martens and Maximilian Werk and Nan Wang and Han Xiao},
  eprint = {2506.18902},
  primaryclass = {cs.AI},
  title = {jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval},
  url = {https://arxiv.org/abs/2506.18902},
  year = {2025},
}

MIRACLVisionRetrieval

Retrieve associated pages according to questions.

Dataset: nvidia/miracl-vision • License: cc-by-sa-4.0 • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	ara, ben, deu, eng, fas, ... (18)	Encyclopaedic	derived	created

Citation

@article{osmulski2025miraclvisionlargemultilingualvisual,
  author = {Radek Osmulski and Gabriel de Souza P. Moreira and Ronay Ak and Mengyao Xu and Benedikt Schifferer and Even Oldridge},
  eprint = {2505.11651},
  journal = {arxiv},
  title = {{MIRACL-VISION: A Large, multilingual, visual document retrieval benchmark}},
  url = {https://arxiv.org/abs/2505.11651},
  year = {2025},
}

Vidore2BioMedicalLecturesRetrieval

Retrieve associated pages according to questions.

Dataset: vidore/biomedical_lectures_v2 • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	deu, eng, fra, spa	Academic	derived	found

Citation

@article{mace2025vidorev2,
  author = {Macé, Quentin and Loison António and Faysse, Manuel},
  journal = {arXiv preprint arXiv:2505.17166},
  title = {ViDoRe Benchmark V2: Raising the Bar for Visual Retrieval},
  year = {2025},
}

Vidore2ESGReportsHLRetrieval

Retrieve associated pages according to questions.

Dataset: vidore/esg_reports_human_labeled_v2 • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@article{mace2025vidorev2,
  author = {Macé, Quentin and Loison António and Faysse, Manuel},
  journal = {arXiv preprint arXiv:2505.17166},
  title = {ViDoRe Benchmark V2: Raising the Bar for Visual Retrieval},
  year = {2025},
}

Vidore2ESGReportsRetrieval

Retrieve associated pages according to questions.

Dataset: vidore/esg_reports_v2 • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	deu, eng, fra, spa	Academic	derived	found

Citation

@article{mace2025vidorev2,
  author = {Macé, Quentin and Loison António and Faysse, Manuel},
  journal = {arXiv preprint arXiv:2505.17166},
  title = {ViDoRe Benchmark V2: Raising the Bar for Visual Retrieval},
  year = {2025},
}

Vidore2EconomicsReportsRetrieval

Retrieve associated pages according to questions.

Dataset: vidore/economics_reports_v2 • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	deu, eng, fra, spa	Academic	derived	found

Citation

@article{mace2025vidorev2,
  author = {Macé, Quentin and Loison António and Faysse, Manuel},
  journal = {arXiv preprint arXiv:2505.17166},
  title = {ViDoRe Benchmark V2: Raising the Bar for Visual Retrieval},
  year = {2025},
}

VidoreArxivQARetrieval

Retrieve associated pages according to questions.

Dataset: vidore/arxivqa_test_subsampled_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@article{faysse2024colpali,
  author = {Faysse, Manuel and Sibille, Hugues and Wu, Tony and Viaud, Gautier and Hudelot, C{\'e}line and Colombo, Pierre},
  journal = {arXiv preprint arXiv:2407.01449},
  title = {ColPali: Efficient Document Retrieval with Vision Language Models},
  year = {2024},
}

VidoreDocVQARetrieval

Retrieve associated pages according to questions.

Dataset: vidore/docvqa_test_subsampled_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@article{faysse2024colpali,
  author = {Faysse, Manuel and Sibille, Hugues and Wu, Tony and Viaud, Gautier and Hudelot, C{\'e}line and Colombo, Pierre},
  journal = {arXiv preprint arXiv:2407.01449},
  title = {ColPali: Efficient Document Retrieval with Vision Language Models},
  year = {2024},
}

VidoreInfoVQARetrieval

Retrieve associated pages according to questions.

Dataset: vidore/infovqa_test_subsampled_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@article{faysse2024colpali,
  author = {Faysse, Manuel and Sibille, Hugues and Wu, Tony and Viaud, Gautier and Hudelot, C{\'e}line and Colombo, Pierre},
  journal = {arXiv preprint arXiv:2407.01449},
  title = {ColPali: Efficient Document Retrieval with Vision Language Models},
  year = {2024},
}

VidoreShiftProjectRetrieval

Retrieve associated pages according to questions.

Dataset: vidore/shiftproject_test_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@article{faysse2024colpali,
  author = {Faysse, Manuel and Sibille, Hugues and Wu, Tony and Viaud, Gautier and Hudelot, C{\'e}line and Colombo, Pierre},
  journal = {arXiv preprint arXiv:2407.01449},
  title = {ColPali: Efficient Document Retrieval with Vision Language Models},
  year = {2024},
}

VidoreSyntheticDocQAAIRetrieval

Retrieve associated pages according to questions.

Dataset: vidore/syntheticDocQA_artificial_intelligence_test_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@article{faysse2024colpali,
  author = {Faysse, Manuel and Sibille, Hugues and Wu, Tony and Viaud, Gautier and Hudelot, C{\'e}line and Colombo, Pierre},
  journal = {arXiv preprint arXiv:2407.01449},
  title = {ColPali: Efficient Document Retrieval with Vision Language Models},
  year = {2024},
}

VidoreSyntheticDocQAEnergyRetrieval

Retrieve associated pages according to questions.

Dataset: vidore/syntheticDocQA_energy_test_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@article{faysse2024colpali,
  author = {Faysse, Manuel and Sibille, Hugues and Wu, Tony and Viaud, Gautier and Hudelot, C{\'e}line and Colombo, Pierre},
  journal = {arXiv preprint arXiv:2407.01449},
  title = {ColPali: Efficient Document Retrieval with Vision Language Models},
  year = {2024},
}

VidoreSyntheticDocQAGovernmentReportsRetrieval

Retrieve associated pages according to questions.

Dataset: vidore/syntheticDocQA_government_reports_test_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@article{faysse2024colpali,
  author = {Faysse, Manuel and Sibille, Hugues and Wu, Tony and Viaud, Gautier and Hudelot, C{\'e}line and Colombo, Pierre},
  journal = {arXiv preprint arXiv:2407.01449},
  title = {ColPali: Efficient Document Retrieval with Vision Language Models},
  year = {2024},
}

VidoreSyntheticDocQAHealthcareIndustryRetrieval

Retrieve associated pages according to questions.

Dataset: vidore/syntheticDocQA_healthcare_industry_test_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@article{faysse2024colpali,
  author = {Faysse, Manuel and Sibille, Hugues and Wu, Tony and Viaud, Gautier and Hudelot, C{\'e}line and Colombo, Pierre},
  journal = {arXiv preprint arXiv:2407.01449},
  title = {ColPali: Efficient Document Retrieval with Vision Language Models},
  year = {2024},
}

VidoreTabfquadRetrieval

Retrieve associated pages according to questions.

Dataset: vidore/tabfquad_test_subsampled_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@article{faysse2024colpali,
  author = {Faysse, Manuel and Sibille, Hugues and Wu, Tony and Viaud, Gautier and Hudelot, C{\'e}line and Colombo, Pierre},
  journal = {arXiv preprint arXiv:2407.01449},
  title = {ColPali: Efficient Document Retrieval with Vision Language Models},
  year = {2024},
}

VidoreTatdqaRetrieval

Retrieve associated pages according to questions.

Dataset: vidore/tatdqa_test_beir • License: mit • Learn more →

Task category	Score	Languages	Domains	Annotations Creators	Sample Creation
text to image (t2i)	ndcg_at_5	eng	Academic	derived	found

Citation

@article{faysse2024colpali,
  author = {Faysse, Manuel and Sibille, Hugues and Wu, Tony and Viaud, Gautier and Hudelot, C{\'e}line and Colombo, Pierre},
  journal = {arXiv preprint arXiv:2407.01449},
  title = {ColPali: Efficient Document Retrieval with Vision Language Models},
  year = {2024},
}