Multimodal LLMs (MLLMs) are the natural extension of large language models to handle multimodal inputs, combining text and image data. They have recently garnered attention due to their capability to address complex tasks involving both modalities. However, their effectiveness is limited to the knowledge acquired during training, which restricts their practical utility. In this work, we introduce a novel method to enhance the adaptability of MLLMs by integrating external knowledge sources. Our proposed model, Reflective LLaVA (ReflectiVA), utilizes reflective tokens to dynamically determine the need for external knowledge and predict the relevance of information retrieved from an external database. Tokens are trained following a two-stage two-model training recipe. This ultimately enables the MLLM to manage external knowledge while preserving fluency and performance on tasks where external knowledge is not needed.

The efficacy of ReflectiVA for knowledge-based visual question answering, highlighting its superior performance compared to existing methods.

In this model space, you will find the Overall Model (stage two) weights of ReflectiVA.

For more information, visit our ReflectiVA repository, our project page and the dataset.

If you make use of our work, please cite our repo:

Paper can be found at https://huggingface.co/papers/2411.16863.