Towards Multimodal Retrieval-Augmented Generation for Medical Visual Question Answering

Beyond Surface Features: Advancing Medical Vision-Language Alignment via Dynamic Evidence-Guided Preference Optimization

Medical large Vision-Language Models (Med-LVLMs) have shown strong potential in multimodal clinical applications such as medical visual question answering and report generation. However, Med-LVLMs remain challenged by hallucinations caused by modality misalignment, where models prioritize textual knowledge over visual evidence and generate outputs that conflict with medical images. To mitigate this issue, recent studies have explored preference optimization to improve image–text alignment, achieving promising results. Despite these advances, existing preference-based meth-ods still face two limitations in medical settings: (1) overfitting to superficial cues, and (2) pseudo convergence of the preference signal. In this paper, we propose Dynamic Evidence-Guided Preference Optimization (DEPO), a new framework that enables evidence-aware and adaptive preference learning for Med-LVLMs. DEPO introduces Multi-Modal Evidence Perturbation (MEP) to suppress non-causal textual and visual shortcuts, and Dispre-ferred Evidence Resampling (DER) to continuously update dispreferred responses as hallucination patterns evolve. Experiments on multiple medical VQA and report generation benchmarks demonstrate consistent improvements over existing methods, with strong robustness across datasets and architectures.

Multimodal AI in healthcare: Review of vision-language foundation models for real-world medical applications.

A definitive taxonomy of the medical VLM landscape is provided, tracing the evolution from early Contrastive Alignment and Generative MLLMs to the cutting-edge frontiers of Dense Pixel-Grounding, Sparse Mixture-of-Experts (MoE), and Reasoning-Incentivized (RL) architectures.

Retrieval-augmented generation for medical question answering: a multi-metric performance evaluation

The proposed framework offers a practical and scalable approach to mitigating hallucinations without requiring task-specific fine-tuning, highlighting the potential of retrieval-augmented approaches for trustworthy artificial intelligence (AI)-assisted healthcare applications.

MedVCoT: Bridging the Modality Gap in Medical VQA Through Latent Visual Reasoning

LocAnyMed: Vision-Language Grounding for Multimodal Medical Images

LocAnyMed-CoT-20K is derived, a rationale-augmented subset that connects anatomical context, visual observations, and spatial conclusions through structured reasoning and further improves cross-source generalization through fine-tuning.

Towards Clinically Faithful Medical Image Captioning via Enhanced Vision-Language Alignment

A framework that strengthens clinical alignment by separating and enhancing training-time alignment and inference-time alignment is proposed and experiments show that complementary visual representations with a multi-encoder design and concept-level auxiliary learning help preserve clinically meaningful information.