This repository contains the data for LOVA3: Learning to Visual Question Answering, Asking and Assessment.
LOVA3 is a framework designed to equip MLLMs with the capabilities to answer, ask, and assess questions in the context of images.
Code:
https://github.com/showlab/LOVA3If you find LOVA3 useful, please cite using this BibTeX:
@inproceedings{
zhao2024lova,
title={{LOVA}3: Learning to Visual Question Answering, Asking and Assessment},
author={Hengyuan… See the full description on the dataset page:
https://huggingface.co/datasets/hhenryz/EvalQABench.
