MOTOR: Multimodal Optimal Transport via Grounded Retrieval in Medical Visual Question Answering
This work addresses retrieval irrelevance in medical VQA, which is crucial for clinical decision-making, but it is incremental as it builds on existing re-ranking methods by adding multimodal context.
The paper tackles the problem of factually incorrect answers in medical visual question answering by proposing MOTOR, a multimodal retrieval and re-ranking approach that uses grounded captions and optimal transport to improve context relevance, achieving an average 6.45% higher accuracy than state-of-the-art methods.
Medical visual question answering (MedVQA) plays a vital role in clinical decision-making by providing contextually rich answers to image-based queries. Although vision-language models (VLMs) are widely used for this task, they often generate factually incorrect answers. Retrieval-augmented generation addresses this challenge by providing information from external sources, but risks retrieving irrelevant context, which can degrade the reasoning capabilities of VLMs. Re-ranking retrievals, as introduced in existing approaches, enhances retrieval relevance by focusing on query-text alignment. However, these approaches neglect the visual or multimodal context, which is particularly crucial for medical diagnosis. We propose MOTOR, a novel multimodal retrieval and re-ranking approach that leverages grounded captions and optimal transport. It captures the underlying relationships between the query and the retrieved context based on textual and visual information. Consequently, our approach identifies more clinically relevant contexts to augment the VLM input. Empirical analysis and human expert evaluation demonstrate that MOTOR achieves higher accuracy on MedVQA datasets, outperforming state-of-the-art methods by an average of 6.45%. Code is available at https://github.com/BioMedIA-MBZUAI/MOTOR.