Datasets › IGLUE

IGLUE (Image-Grounded Language Understanding Evaluation)

Introduced by Emanuele Bugliarello et al. in IGLUE: A Benchmark for Transfer Learning across Modalities, Tasks, and Languages27 Jan 2022 archive 2025-07-28

The Image-Grounded Language Understanding Evaluation (IGLUE) benchmark brings together—by both aggregating pre-existing datasets and creating new ones—visual question answering, cross-modal retrieval, grounded reasoning, and grounded entailment tasks across 20 diverse languages. The benchmark enables the evaluation of multilingual multimodal models for transfer learning, not only in a zero-shot setting, but also in newly defined few-shot learning setups.

Homepage JSON

Benchmarks archive 2025-07-28

No leaderboard in the archive resolves to this dataset.

Papers archive 2025-07-28

No paper in the archive has a leaderboard row on this dataset; the archive counts 31 papers for it but never published that list.

Dataset loaders archive 2025-07-28

No loader listed in the archive.

Tasks archive 2025-07-28

Visual Reasoning Zero-Shot Cross-Lingual Transfer Zero-Shot Cross-Lingual Text-to-Image Retrieval Zero-Shot Cross-Lingual Image-to-Text Retrieval Zero-Shot Cross-Lingual Visual Natural Language Inference Zero-Shot Cross-Lingual Visual Reasoning Zero-Shot Cross-Lingual Visual Question Answering Max-Shot Cross-Lingual Visual Natural Language Inference Max-Shot Cross-Lingual Visual Question Answering Max-Shot Cross-Lingual Visual Reasoning Max-Shot Cross-Lingual Text-to-Image Retrieval Max-Shot Cross-Lingual Image-to-Text Retrieval

License archive 2025-07-28

Unknown

Modalities archive 2025-07-28

Images Texts

Languages archive 2025-07-28

English French Spanish German Chinese Bengali Japanese Russian Portuguese Arabic Bulgarian Danish Estonian Indonesian Korean Tamil Turkish Vietnamese Greek Swahili

Variants archive 2025-07-28

MaRVL
IGLUE
XVNLI
xGQA
xFlickr&CO
WIT (IGLUE)

6 variant names, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections