RS-MoE: A Vision–Language Model With Mixture of Experts for Remote Sensing Image Captioning and Visual Question Answering

Remote sensing image captioning (RSIC) presents unique challenges and plays a critical role in applications such as environmental monitoring, urban planning, and disaster management. Traditional RSIC methods often struggle to produce rich and diverse descriptions. Recently, with significant advancements in vision-language models (VLMs), efforts have emerged to integrate these models into the remote sensing domain and to introduce richly descriptive datasets specifically designed to enhance VLM training. However, most current RSIC models generally apply only fine-tuning to these datasets without developing models tailored to the unique characteristics of remote sensing imagery. This article proposes RS-MoE, the first mixture of expert (MoE)-based VLM specifically customized for remote sensing domain. Unlike traditional MoE models, the core of RS-MoE is the MoE block, which incorporates a novel instruction router and multiple lightweight large language models (LLMs) as expert models. The instruction router is designed to generate specific prompts tailored for each corresponding LLM, guiding them to focus on distinct aspects of the RSIC task. This design not only allows each expert LLM to concentrate on a specific subset of the task, thereby enhancing the specificity and accuracy of the generated captions, but also improves the scalability of the model by facilitating parallel processing of subtasks. In addition, we present a two-stage training strategy for tuning our RS-MoE model to prevent performance degradation due to sparsity. We fine-tuned our model on the RSICap dataset using our proposed training strategy. Experimental results on the RSICap dataset, along with evaluations on other traditional datasets where no additional fine-tuning was applied, demonstrate that our model achieves state-of-the-art performance in generating precise and contextually relevant captions. Notably, our RS-MoE-1B variant achieves performance comparable to 13B VLMs, demonstrating the efficiency of our model design. Moreover, our model demonstrates promising generalization capabilities by consistently achieving state-of-the-art performance on the remote sensing visual question answering (RSVQA) task.