Co-VQA : Answering by Interactive Sub Question Sequence
This addresses the problem of handling complex visual questions for AI systems by introducing an interactive, interpretable approach, though it is incremental in building on existing VQA methods.
The paper tackles Visual Question Answering (VQA) by simulating human-like decomposition of complex questions into sub-question sequences, proposing a Co-VQA framework with Questioner, Oracle, and Answerer components. It achieves state-of-the-art results on the VQA-CP v2 dataset, demonstrating improved interpretability and error traceability.
Most existing approaches to Visual Question Answering (VQA) answer questions directly, however, people usually decompose a complex question into a sequence of simple sub questions and finally obtain the answer to the original question after answering the sub question sequence(SQS). By simulating the process, this paper proposes a conversation-based VQA (Co-VQA) framework, which consists of three components: Questioner, Oracle, and Answerer. Questioner raises the sub questions using an extending HRED model, and Oracle answers them one-by-one. An Adaptive Chain Visual Reasoning Model (ACVRM) for Answerer is also proposed, where the question-answer pair is used to update the visual representation sequentially. To perform supervised learning for each model, we introduce a well-designed method to build a SQS for each question on VQA 2.0 and VQA-CP v2 datasets. Experimental results show that our method achieves state-of-the-art on VQA-CP v2. Further analyses show that SQSs help build direct semantic connections between questions and images, provide question-adaptive variable-length reasoning chains, and with explicit interpretability as well as error traceability.