Language Plays a Pivotal Role in the Object-Attribute Compositional Generalization of CLIP

Autor:	Abbasi, Reza, Samiei, Mohammad, Rohban, Mohammad Hossein, Baghshah, Mahdieh Soleymani
Rok vydání:	2024
Předmět:	Computer Science - Computer Vision and Pattern Recognition Computer Science - Computation and Language Computer Science - Machine Learning
Druh dokumentu:	Working Paper
Popis:	Vision-language models, such as CLIP, have shown promising Out-of-Distribution (OoD) generalization under various types of distribution shifts. Recent studies attempted to investigate the leading cause of this capability. In this work, we follow the same path, but focus on a specific type of OoD data - images with novel compositions of attribute-object pairs - and study whether such models can successfully classify those images into composition classes. We carefully designed an authentic image test dataset called ImageNet-AO, consisting of attributes for objects that are unlikely encountered in the CLIP training sets. We found that CLIPs trained with large datasets such as OpenAI CLIP, LAION-400M, and LAION-2B show orders-of-magnitude improvement in effective compositional OoD generalization compared to both supervised models and CLIPs trained with smaller datasets, such as CC-12M and YFCC-15M. Our results provide evidence that the scale and diversity of training data and language supervision play a key role in unlocking the compositional generalization abilities of vision-language models. Comment: Oral accepted at OODCV 2023(http://www.ood-cv.org)
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2403.18525 Zobrazit plný text záznamu View this record from Arxiv