In the rapidly evolving landscape of artificial intelligence, multi-modal large language models are emerging as a significant area of interest. These models, which combine various forms of data input, are becoming increasingly popular. However, understanding their internal mechanisms remains a complex task. Numerous advancements have been made in the field of explainability tools and mechanisms, yet there is still much to explore. In this work, we present a novel interactive application aimed towards understanding the internal mechanisms of large vision-language models. Our interface is designed to enhance the interpretability of the image patches, which are instrumental in generating an answer, and assess the efficacy of the language model in grounding its output in the image. With our application, a user can systematically investigate the model and uncover system limitations, paving the way for enhancements in system capabilities. Finally, we present a case study of how our application can aid in understanding failure mechanisms in a popular large multi-modal model: LLaVA.
@article{arxiv.2404.03118,
title = {LVLM-Interpret: An Interpretability Tool for Large Vision-Language Models},
author = {Gabriela Ben Melech Stan and Estelle Aflalo and Raanan Yehezkel Rohekar and Anahita Bhiwandiwalla and Shao-Yen Tseng and Matthew Lyle Olson and Yaniv Gurwicz and Chenfei Wu and Nan Duan and Vasudev Lal},
journal= {arXiv preprint arXiv:2404.03118},
year = {2024}
}