- Enables seamless use of internal documents, supports automatic updates of RAG agents -

Fujitsu today officially released v3.3 of its proprietary AI assistant building platform, Super RAG™ (Super Lag), which leverages complex internal documents across a wide range of processes to provide highly accurate knowledge search and answers using generative AI.

In this update, we have officially released a feature to integrate with Microsoft's SharePoint, a key feature that will powerfully support improving corporate productivity. This provides an environment where users can safely and easily build and operate their own company-specific AI assistant through SharePoint integration, even without specialized knowledge of RAG*. This will make internal knowledge utilization and information sharing more efficient than ever before, powerfully promoting digital transformation in companies.

*Retrieval Augmented Generation technology that adds unique information, such as internal documents, to large-scale language models

◇ "Super RAG" v3.3 Update: Main features of SharePoint integration

With this enhancement, even users without specialized knowledge can easily build and operate their own company-specific AI assistant through SharePoint integration, further accelerating the utilization of knowledge and information sharing within the company.

1. Flexible collaboration management on a site-by-site basis

Connection settings with "Super RAG" can be made for each SharePoint site. In addition, you can select or exclude any document within a site, allowing flexible collaboration management according to your company's operational policies, such as viewing and editing by department.

2. Supports two types of updates: automatic and manual

In addition to automatic updates every 24 hours between "Super RAG" and SharePoint, administrators can also manually update documents. When documents are updated, only the changed parts are imported, ensuring that the latest information is always quickly reflected in "Super RAG's" AI assistant. By efficiently updating information, "Super RAG" can always provide answers based on the latest and most accurate information.

3. Azure ADSecure connection with SSO integration

The authentication infrastructure of "Super RAG" now supports "Azure Active Directory (Azure AD)." Utilizing the advanced authentication features adopted by many companies makes it easier to meet internal security standards and compliance requirements. Authentication integration with Azure AD also enables centralized management of user accounts, reducing the operational burden on IT departments while achieving safe and efficient integration.

◇ Main effects of implementation

- Generate answers that reflect the latest documents on SharePoint
・ Streamline knowledge sharing across departments and projects
・ Promoting "knowledge discovery" by RAG agents
- Reduces the burden on IT departments and enables safe collaboration

◇ Future developments of "Super RAG"

Cinnamon AI will continue to expand the functionality of "Super RAG" to expand the possibilities of knowledge utilization in companies. In the next version, we plan to add a function to integrate with Box.

■ "Super RAG™" is highly accurate and requires no tuning, and can handle complex documents including charts and graphs

Cinnamon AI's "Super RAG" is a proprietary AI product that allows you to build LLMs that analyze complex documents such as tables and figures and generate highly accurate answers without any tuning. It uses proprietary document analysis technology to perform advanced analysis of unstructured data, which is said to account for 80% of data held by companies and is unsuitable for data utilization, and enables optimal structuring according to content type*, enabling business automation through the generation of LLM answers using the RAG system. It provides a one-stop solution that enables highly accurate searches using a graph database that utilizes the relationships between content units within documents, and connectivity to various systems and applications through a simple UI and API.

"Super RAG" can be implemented in a wide range of business processes. For example, you can easily import large amounts of specialized documents such as financial statements, technical information, and internal regulations, and LLM can quickly generate appropriate answers*. It can also be used for tasks that require specialized knowledge, such as responding to inquiries, creating business plans, writing reports, and predicting incidents, as well as small-volume, high-mix operations that require tacit knowledge. *Response accuracy may vary depending on the type of document.

Complex documents including figures and tables that can be used with "Super RAG" (image)

■ Related pages

Cinnamon AI "Super RAG": https://cinnamon.ai/super-rag/
Cinnamon AI "Super RAG" materials are available here Click here Please contact us at:

*"Super RAG™" mentioned in this press release is a trademark of Cinnamon Co., Ltd.