Demos
No-Code Tool Enables Domain Experts in High-Compliance Settings to Bootstrap Task-Specific Models with Human-in-the-Loop Workflows
John Snow Labs, the AI for healthcare company, today announced the release of its Generative AI Lab, a first-of-its-kind tool that enables healthcare domain experts to train, tune, test, and share models without coding or involving data scientists. The Generative AI Lab is the only no-code solution designed for the most commonly requested use cases since the emergence of large language models (LLMs).
One such use case is the ability to use LLMs to bootstrap smaller, task-specific models. In other words, domain experts can provide zero-shot prompts, see how they perform on real documents, and provide feedback to improve accuracy, which is used to train models downstream. In addition to the accuracy boost, this also results in small models which are far cheaper to run at scale. Instead of buying massive GPUs, enterprises can leverage generative AI efficiently, without paying per token for inference, and all without a team of data scientists on staff.
Built specifically for high-compliance environments, the Generative AI Lab can run entirely behind an organization’s firewall, with no need to call a third party API, or even have Internet access. Users can organize and share models, prompts, and rules within one private enterprise hub. This central repository enables teams to securely share, search, filter, test, publish, import, and export an organization’s private models, prompts, and rules, with a full range of built-in security features.
The Generative AI Lab also supports human-in-the-loop workflows. In industries like healthcare, in which regulatory-grade accuracy is a requirement, human validation of model results is often a critical requirement. While models handle the legwork, the Generative AI Lab not only augments decision-making abilities for domain experts, but empowers them to train better models as they go. The tool supports task management, full audit trails, custom review and approval workflows, versioning, and analytics, fully supporting the human-in-the-loop needs of high-compliance industries.
“We’ve been using the Generative AI Lab for over a year to extract information from safety data sheets with regulatory-grade accuracy,” said Vishnu Vettrivel, CEO, Wisecube AI. “It’s utterly unique in combining three capabilities—starting quickly from existing AI models without coding; an easy-to-use interface for human-in-the-loop workflows; and a feedback loop that makes the models better as you go. No other tool we’ve looked at comes close.”
“The Generative AI Lab was built to support the need for small, fine-tuned healthcare-specific models, and human-in-the-loop workflows to improve them,” said David Talby, CTO, John Snow Labs. “As a no-code tool, the Generative AI Lab empowers domain experts, like doctors and lawyers, to maximize the value generative AI can provide, ultimately leading to better experiences in healthcare and other high-compliance environments.”
An overview of the Generative AI Lab will be presented at this year’s Healthcare NLP Summit in a keynote session titled, “Using LLMs to Build Task-Specific AI Models with the No-Code Generative AI Lab.” The free event, being held April 2-3, is the world’s largest gathering for the healthcare AI and applied NLP community. You can register here.
Follow John Snow Labs and #NLPSummit on LinkedIn for the latest news and updates.
