Scaffolding Coordinates to Promote Vision-Language Coordination in Large Multi-Modal Models
This addresses the need for better vision-language coordination in LMMs for challenging tasks, representing an incremental improvement over existing prompting techniques.
The paper tackles the problem of limited performance in Large Multi-Modal Models (LMMs) for complex vision-language reasoning by proposing Scaffold prompting, which overlays a dot matrix and uses coordinates as references, showing superiority over GPT-4V with textual CoT prompting in extensive experiments.
State-of-the-art Large Multi-Modal Models (LMMs) have demonstrated exceptional capabilities in vision-language tasks. Despite their advanced functionalities, the performances of LMMs are still limited in challenging scenarios that require complex reasoning with multiple levels of visual information. Existing prompting techniques for LMMs focus on either improving textual reasoning or leveraging tools for image preprocessing, lacking a simple and general visual prompting scheme to promote vision-language coordination in LMMs. In this work, we propose Scaffold prompting that scaffolds coordinates to promote vision-language coordination. Specifically, Scaffold overlays a dot matrix within the image as visual information anchors and leverages multi-dimensional coordinates as textual positional references. Extensive experiments on a wide range of challenging vision-language tasks demonstrate the superiority of Scaffold over GPT-4V with the textual CoT prompting. Our code is released in https://github.com/leixy20/Scaffold.