Task
Visual Question Answering
M
wudq/MVEI_PLUS
Visual Question Answering · wudq· 36.4K
Unknown
18 GB
text Original MVEI (Hugging Face: wudq/MVEI) ── Evaluation (GitHub: wdqqdw/EmObserver) ── Inference Results (this repository)
# download instantly
mlforge datasets pull wudq/MVEI_PLUS
mlforge datasets pull wudq/MVEI_PLUS
Dataset details
Language
en
License
Unknown
Size
18 GB
Creator
wudq
Downloads
36.4K
Source
huggingface_datasets
Updated
2026-07-30
About wudq/MVEI_PLUS
Resource Role Link --- --- --- Original conference release Original MVEI code and project material wdqqdw/MVEI Original MVEI dataset Source MVEI benchmark release wudq/MVEI Original INSETS-462k dataset Source INSETS-462k release wudq/INSETS-462k MVEIPLUS This expanded evaluation and training-data release wudq/MVEIPLUS Public code Inference, evaluation, and four-stage training pipeline wdqqdw/EmObserver EmObserver Released model checkpoint evaluated by the packaged baseline outputs wudq/EmObserver