ReasonVQA: A Multi-hop Reasoning Benchmark with Structural Knowledge for Visual Question Answering
This provides a benchmark for advancing VQA by addressing multi-hop reasoning with structural knowledge, though it is incremental as it builds on existing VQA tasks.
The authors introduced ReasonVQA, a new dataset for Visual Question Answering that integrates structured encyclopedic knowledge to generate complex, multi-hop questions, and found that it significantly challenges state-of-the-art VQA models. The dataset is scalable, with the current version being over ten times larger than existing knowledge-based VQA datasets.
In this paper, we propose a new dataset, ReasonVQA, for the Visual Question Answering (VQA) task. Our dataset is automatically integrated with structured encyclopedic knowledge and constructed using a low-cost framework, which is capable of generating complex, multi-hop questions. We evaluated state-of-the-art VQA models on ReasonVQA, and the empirical results demonstrate that ReasonVQA poses significant challenges to these models, highlighting its potential for benchmarking and advancing the field of VQA. Additionally, our dataset can be easily scaled with respect to input images; the current version surpasses the largest existing datasets requiring external knowledge by more than an order of magnitude.