ViVQA

ViVQA is a dataset designed for Vietnamese Visual Question Answering that contains visual and textual information to evaluate models' ability to integrate linguistic and visual features in the Vietnamese context.

Papers using ViVQA (3)

AutoViVQA: A Large-Scale Automatically Constructed Dataset for Vietnamese Visual Question Answering2026Advancing Vietnamese Visual Question Answering with Transformer and Convolutional Integration2024 · 7 citesPAT: Parallel Attention Transformer for Visual Question Answering in Vietnamese2023