Neuro-Symbolic VQA: A review from the perspective of AGI desiderata

TL;DR

This paper reviews neuro-symbolic approaches to visual question answering (VQA) in the context of artificial general intelligence (AGI) goals, analyzing their strengths, limitations, and future potential.

Contribution

It provides a comprehensive analysis of neuro-symbolic VQA systems from the perspective of AGI desiderata, highlighting challenges and opportunities for future development.

Findings

Neuro-symbolic VQA systems exhibit promising AGI-like properties.

Current benchmarks may not fully capture the capabilities of these systems.

The paper discusses how different properties of models align or conflict with AGI goals.

Abstract

An ultimate goal of the AI and ML fields is artificial general intelligence (AGI); although such systems remain science fiction, various models exhibit aspects of AGI. In this work, we look at neuro-symbolic (NS)approaches to visual question answering (VQA) from the perspective of AGI desiderata. We see how well these systems meet these desiderata, and how the desiderata often pull the scientist in opposing directions. It is my hope that through this work we can temper model evaluation on benchmarks with a discussion of the properties of these systems and their potential for future extension.