Coarse-to-Fine Contrastive Learning in Image-Text-Graph Space for Improved Vision-Language Compositionality
Submission Type: Regular Long Paper
Submission Track: Language Grounding to Vision, Robotics and Beyond
Keywords: Vision-Language Compositionality, Systematic Generalization, Vision-Language Contrastive Learning, Multimodal Foundation Models
TL;DR: Significant improvements in compositional reasoning capabilities of Vision Language models via a new scene-graph guided coarse-to-fine contrastive learning technique.
Abstract: Contrastively trained vision-language models have achieved remarkable progress in vision and language representation learning. However, recent research has highlighted severe limitations of these models in their ability to perform compositional reasoning over objects, attributes, and relations. Scene graphs have emerged as an effective way to understand images compositionally. These are graph-structured semantic representations of images that contain objects, their attributes, and relations with other objects in a scene. In this work, we consider the scene graph parsed from text as a proxy for the image scene graph and propose a graph decomposition and augmentation framework along with a coarse-to-fine contrastive learning objective between images and text that aligns sentences of various complexities to the same image. We also introduce novel negative mining techniques in the scene graph space for improving attribute binding and relation understanding. Through extensive experiments, we demonstrate the effectiveness of our approach that significantly improves attribute binding, relation understanding, systematic generalization, and productivity on multiple recently proposed benchmarks (For example, improvements
up to $\mathbf{18}$% for systematic generalization, $\mathbf{16.5}$% for relation understanding over a strong baseline), while achieving similar or better performance than CLIP on various general multimodal tasks.
Submission Number: 78
Loading