Multimodal Hypothetical Summary for Retrieval-based Multi-image Question Answering
This addresses a specific bottleneck in multi-modal AI systems for researchers and practitioners, though it is incremental in nature.
The paper tackles the problem of cascading errors in retrieval-based multi-image question answering by proposing a method that uses multimodal hypothetical summaries to transform retrieval into text-to-text, achieving a 3.7% absolute improvement over state-of-the-art methods on RETVQA and a 14.5% improvement over CLIP.
Retrieval-based multi-image question answering (QA) task involves retrieving multiple question-related images and synthesizing these images to generate an answer. Conventional "retrieve-then-answer" pipelines often suffer from cascading errors because the training objective of QA fails to optimize the retrieval stage. To address this issue, we propose a novel method to effectively introduce and reference retrieved information into the QA. Given the image set to be retrieved, we employ a multimodal large language model (visual perspective) and a large language model (textual perspective) to obtain multimodal hypothetical summary in question-form and description-form. By combining visual and textual perspectives, MHyS captures image content more specifically and replaces real images in retrieval, which eliminates the modality gap by transforming into text-to-text retrieval and helps improve retrieval. To more advantageously introduce retrieval with QA, we employ contrastive learning to align queries (questions) with MHyS. Moreover, we propose a coarse-to-fine strategy for calculating both sentence-level and word-level similarity scores, to further enhance retrieval and filter out irrelevant details. Our approach achieves a 3.7% absolute improvement over state-of-the-art methods on RETVQA and a 14.5% improvement over CLIP. Comprehensive experiments and detailed ablation studies demonstrate the superiority of our method.