# GQA: A New Dataset for Real-World Visual Reasoning and Compositional Question Answering

## TL;DR

GQA introduces a large, diverse dataset with a robust question generation engine and new evaluation metrics, aiming to advance real-world visual reasoning and compositional question answering in AI models.

## Contribution

The paper presents GQA, a comprehensive dataset with 22 million questions, a novel question engine leveraging scene graphs, and new metrics for evaluating visual reasoning models.

## Key findings

- Baseline models achieve 42.1% accuracy, while state-of-the-art models reach 54.1%.
- Human performance on GQA is 89.3%.
- The dataset enables detailed analysis of model reasoning capabilities.

## Abstract

We introduce GQA, a new dataset for real-world visual reasoning and compositional question answering, seeking to address key shortcomings of previous VQA datasets. We have developed a strong and robust question engine that leverages scene graph structures to create 22M diverse reasoning questions, all come with functional programs that represent their semantics. We use the programs to gain tight control over the answer distribution and present a new tunable smoothing technique to mitigate question biases. Accompanying the dataset is a suite of new metrics that evaluate essential qualities such as consistency, grounding and plausibility. An extensive analysis is performed for baselines as well as state-of-the-art models, providing fine-grained results for different question types and topologies. Whereas a blind LSTM obtains mere 42.1%, and strong VQA models achieve 54.1%, human performance tops at 89.3%, offering ample opportunity for new research to explore. We strongly hope GQA will provide an enabling resource for the next generation of models with enhanced robustness, improved consistency, and deeper semantic understanding for images and language.