VTQA: Visual Text Question Answering via Entity Alignment and Cross-Media Reasoning

TL;DR

This paper introduces VTQA, a new visual text question answering challenge that emphasizes entity alignment, multi-hop reasoning, and open-ended answer generation across vision and language modalities.

Contribution

It presents a novel dataset and task for multimedia entity alignment and reasoning, advancing beyond traditional VQA benchmarks.

Findings

New dataset with 23,781 questions based on 10,124 image-text pairs

Emphasizes multi-hop reasoning and entity alignment across modalities

Supports open-ended answer generation in VQA tasks

Abstract

The ideal form of Visual Question Answering requires understanding, grounding and reasoning in the joint space of vision and language and serves as a proxy for the AI task of scene understanding. However, most existing VQA benchmarks are limited to just picking the answer from a pre-defined set of options and lack attention to text. We present a new challenge with a dataset that contains 23,781 questions based on 10124 image-text pairs. Specifically, the task requires the model to align multimedia representations of the same entity to implement multi-hop reasoning between image and text and finally use natural language to answer the question. The aim of this challenge is to develop and benchmark models that are capable of multimedia entity alignment, multi-step reasoning and open-ended answer generation.