CLIPPO: Image-and-Language Understanding from Pixels Only Article Swipe

MAKE IMPACT

Michael Tschannen , Basil Mustafa , Neil Houlsby · YOU? · · 2022 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2212.08045

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.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2212.08045

PDF

https://arxiv.org/pdf/2212.08045

OA Status

green

Cited By

1

Related Works

10

OpenAlex ID

https://openalex.org/W4311730904

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4311730904

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2212.08045

Digital Object Identifier

Title

CLIPPO: Image-and-Language Understanding from Pixels Only

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2022

Year of publication

Publication date

2022-12-15

Full publication date if available

Authors

Michael Tschannen, Basil Mustafa, Neil Houlsby

List of authors in order

Landing page

https://arxiv.org/abs/2212.08045

Publisher landing page

PDF URL

https://arxiv.org/pdf/2212.08045

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2212.08045

Direct OA link when available

Concepts

Computer science, Artificial intelligence, Encoder, Rendering (computer graphics), Natural language processing, Pixel, Language model, Sentence, Embedding, Transformer, Exploit, Natural language, Computer security, Physics, Voltage, Operating system, Quantum mechanics

Top concepts (fields/topics) attached by OpenAlex

Cited by

1

Total citation count in OpenAlex

Citations by year (recent)

2023: 1

Per-year citation counts (last 5 years)

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload