datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
3D-Object-Conflictmultilingual-crossmodal-conflict-3D_Objects
Multilingual Cross-Modal Conflict — 3D Objects
A multilingual counterfactual MCQ dataset built from rendered 3D object scenes.
Each row contains a rendered 3D scene image, two captions (original vs counterfactual), and a multiple-choice question probing whether a VLM follows the image or the misleading text.
Languages
Language
Code
Rows
English
en
150
Hindi
hi
150
Telugu
te
150
Bahasa Indonesia
id
150
Columns
Column
Type… See the full description on the dataset page: https://huggingface.co/datasets/apart-global-south-hack/multilingual-crossmodal-conflict-3D_Objects.