CLIPPO: Image-and-Language Understanding from Pixels Only

TL;DR

CLIPPO introduces a unified pixel-based model that performs image, text, and multimodal tasks using only contrastive loss, achieving competitive results with fewer parameters and no text-specific components.

Contribution

The paper presents CLIPPO, a novel pixel-only multimodal model that unifies image and text processing without task-specific modules or tokenizers, simplifying training and inference.

Findings

Performs image retrieval and zero-shot classification nearly as well as CLIP.

Achieves strong natural language understanding without word-level training.

Excels in multilingual multimodal retrieval without modifications.

Abstract

Multimodal models are becoming increasingly effective, in part due to unified components, such as the Transformer architecture. However, multimodal models still often consist of many task- and modality-specific pieces and training procedures. For example, CLIP (Radford et al., 2021) trains independent text and image towers via a contrastive loss. We explore an additional unification: the use of a pure pixel-based model to perform image, text, and multimodal tasks. Our model is trained with contrastive loss alone, so we call it CLIP-Pixels Only (CLIPPO). CLIPPO uses a single encoder that processes both regular images and text rendered as images. CLIPPO performs image-based tasks such as retrieval and zero-shot image classification almost as well as CLIP-style models, with half the number of parameters and no text-specific tower or embedding. When trained jointly via image-text contrastive learning and next-sentence contrastive learning, CLIPPO can perform well on natural language understanding tasks, without any word-level loss (language modelling or masked language modelling), outperforming pixel-based prior work. Surprisingly, CLIPPO can obtain good accuracy in visual question answering, simply by rendering the question and image together. Finally, we exploit the fact that CLIPPO does not require a tokenizer to show that it can achieve strong performance on multilingual multimodal retrieval without modifications.