Graph-of-Mark: Promote Spatial Reasoning in Multimodal Language Models with Graph-Based Visual Prompting

1. 1Partition the input image into object regions
2. 2Annotate the regions with graph-based marks
3. 3Feed the augmented image to the multimodal language model
4. 4Evaluate the model's spatial reasoning capabilities

AI engineers and ML researchers can benefit from this approach to improve the performance of multimodal language models, especially in tasks that require spatial reasoning and visual understanding

💡 Graph-based visual prompting can improve the spatial reasoning capabilities of multimodal language models

💡 Graph-of-Mark: Enhancing spatial reasoning in multimodal language models with graph-based visual prompting