FA864924P0396

Small Technology Transfer Research Program (STTR) Phase I

Streamlining Acquisition Operations with Generative AI-powered Natural Language Search

Kyndi has solved the problem of finding answers in unstructured text by successfully merging Deep Learning Natural Language Processing (NLP) and Symbolic AI into a new AI paradigm called Neuro-Symbolic Language Models (NSLMs). NLSMs overcome the limitations of traditional supervised machine learning systems. The result is an AI platform that is accessible to non-technical people and which can find answers in one-click, but does not require labeled training data or skilled AI engineers, which equates to operational effectiveness (95%+ reduction in analysis time) in weeks, not months or quarters. Implementations only require a list of domain-specific terms to be identified and propagated into the base deep learning large language model. Kyndi leverages some of the same underlying technologies as ChatGPT and can therefore find answers to questions, but Kyndi returns accurate and trustworthy answers verbatim instead of generating a custom response, which will be wrong an appreciable percentage of the time.

The goal of phase I is to establish the technical merit, feasibility, and commercial potential of proposed R&D efforts and determine the quality of performance of the small business awardee organization. STTRs are completed in conjunction with a research institution.

University of Notre Dame

AFX23D-TCSO1

FX23D-TCSO1-0294

X23.D

Jenna Roeche