Recent years have seen remarkable development in the creation of sophisticated language-vision models. Real-world applications rely heavily on multimodal material, particularly language-vision data, which includes texts, photos, and videos. 

However, domain knowledge is required for training and evaluating these models across tasks and datasets, and they are not necessarily open to new researchers and practitioners. This is primarily because preparing the necessary experiment setup is a lot of work and is time-consuming regardless of the model, dataset, or task evaluation being used.

Salesforce researchers have developed LAVIS (short for LAnguage-VISion), an open-source library for training and evaluating state-of-the-art language-vision models on a rich family of common tasks and datasets and for off-the-shelf inference on customized language-vision data. This will make the emerging language-vision intelligence and capabilities available to a wider audience, encourage practical adoption, and reduce repetitive efforts in future development.

LAVIS is an all-inclusive, modular, and future-proof language-vision library that works with standard tasks, data sets, and cutting-edge models. LAVIS’s overarching goal is to offer data scientists, machine learning engineers, and academics a streamlined means to examine, troubleshoot, and clarify their multimodal data.

The LAVIS features that are most notable are:

According to the team, extending the library’s current selection of language-vision models, jobs, and datasets is a top priority for future releases. They also intend to provide greater parallelism support for scalable training and inference.

Asif Razzaq is an AI Journalist and Cofounder of Marktechpost, LLC. He is a visionary, entrepreneur and engineer who aspires to use the power of Artificial Intelligence for good.

Asif's latest venture is the development of an Artificial Intelligence Media Platform (Marktechpost) that will revolutionize how people can find relevant news related to Artificial Intelligence, Data Science and Machine Learning.

Asif was featured by Onalytica in it’s ‘Who’s Who in AI? (Influential Voices & Brands)’ as one of the 'Influential Journalists in AI' (https://onalytica.com/wp-content/uploads/2021/09/Whos-Who-In-AI.pdf). His interview was also featured by Onalytica (https://onalytica.com/blog/posts/interview-with-asif-razzaq/).

Free-2 Min AI Newsletter Join Our AI Community EmailEnter your email address Subscribe

Marktechpost is a California based AI News Platform providing easy-to-consume, byte size updates in machine learning, deep learning, and data science research

© 2021 Marktechpost LLC. All Rights Reserved. Made with ❤️ in California

Learn the AI best practices from 120+ experts at the TransformX Conference