← all datasets

EndoVis-18-VQA

Emerging
3papers using it
2025first seen

The 'EndoVis-18-VQA' is a dataset that contains video question-answer pairs specifically designed to evaluate multi-step reasoning in surgical video question answering.

🔎 Find this dataset

Papers using EndoVis-18-VQA (2)

EndoVis-18-VQA — datasets — multimodal