> ## Documentation Index
> Fetch the complete documentation index at: https://docs.omnifact.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Uploaded Files Retrieval

> Understand how Spaces use Uploaded Files to provide AI responses based on your organization's documents

export const Glossary = ({children, term}) => {
  const glossaryDefinitions = {
    'ai': 'Artificial Intelligence - computer systems that can perform tasks typically requiring human intelligence',
    'artificial-intelligence-ai': 'Computer systems that can perform tasks typically requiring human intelligence',
    'ai-assistant': 'An AI system designed to help users with tasks through conversation',
    'space': 'A dedicated environment with a specialized AI assistant and optional Uploaded Files',
    'spaces': 'Dedicated environments with specialized AI assistants and optional Uploaded Files',
    'knowledge': 'The sub-menu in Space settings where you manage Uploaded Files and Connected Sources',
    'uploaded-files': 'Documents manually uploaded to a Space under the Knowledge tab',
    'prompt': 'Instructions or questions you give to an AI assistant',
    'response': 'The AI assistant\'s answer to your prompt or question',
    'query': 'A question or request for information',
    'context': 'Background information that helps the AI understand your request',
    'llm': 'Large Language Model - an AI system trained on vast amounts of text data to understand and generate human language',
    'large-language-model-llm': 'An AI system trained on vast amounts of text data to understand and generate human language',
    'privacy-filter': 'A system that automatically detects and masks sensitive information before it reaches AI models, and restricts images from being sent to non-EU hosted models for compliance',
    'chat-instructions': 'Custom guidelines that define how an AI assistant behaves and responds in a Space',
    'pattern-matching': 'The process by which AI recognizes and uses patterns in data to provide responses',
    'patterns': 'Recognizable structures or trends in data that AI systems can identify and use',
    'training-data': 'The text and information used to teach an AI system how to understand and respond to requests',
    'natural-language': 'Everyday human language that people use to communicate, as opposed to computer code or formal syntax',
    'hallucination': 'When an AI generates information that sounds plausible but is actually incorrect or fabricated',
    'filtering': 'The process of automatically detecting and temporarily replacing sensitive information to protect it during AI processing',
    'masking': 'Temporarily replacing sensitive information with generic labels to protect it from being seen by external AI models',
    're-insertion': 'The automatic process of putting original sensitive information back into AI responses after they\'ve been generated with filtered data',
    'retrieval': 'The process of finding and accessing relevant documents from Uploaded Files to answer questions',
    'hosting': 'Where an AI model is deployed and made available for use - either by third-party providers, on your organization\'s own servers, or in private cloud environments',
    'cite': 'To reference or mention the source of information, helping you know where an AI\'s answer came from',
    'sidebar': 'A panel that appears on the side of the screen showing additional information and options',
    'toggle': 'A button or control that switches something on or off, like opening and closing a sidebar',
    'navigation': 'The system of menus and buttons that help you move between different parts of Omnifact',
    'favoriting': 'Marking a Space as a favorite to pin it to the main navigation for quick access',
    'pinning': 'Attaching a Space to the main navigation bar so it\'s always visible and easily accessible'
  };
  const definition = glossaryDefinitions[term];
  return <Tooltip tip={definition}>{children}</Tooltip>;
};

Uploaded Files <Glossary term="retrieval">retrieval</Glossary> is a powerful capability that allows the <Glossary term="ai-assistant">AI assistant</Glossary> within a <Glossary term="space">Space</Glossary> to access and leverage your organization's documents when responding to your questions. This feature transforms generic AI responses into contextual, organization-specific answers grounded in your proprietary information.

## What is Uploaded Files?

A Uploaded Files in Omnifact is a curated collection of documents that are accessible to a Space's AI assistant. It serves as the specialized knowledge foundation for each Space, providing:

<Note>
  If your organization has enabled cloud storage integrations, your Space might also include [Connected Sources](/en/platform/core-features/spaces/connected-sources) which function seamlessly alongside manually uploaded files for retrieval.
</Note>

<CardGroup cols={2}>
  <Card title="Document Repository" icon="folder-open">
    A secure storage location for important files relevant to the Space's purpose or domain.
  </Card>

  <Card title="Contextual Intelligence" icon="brain">
    Source material that enhances the AI assistant's ability to provide relevant, accurate, and organization-specific information.
  </Card>

  <Card title="Proprietary Uploaded Files" icon="lock">
    A way to leverage your internal documents that aren't available to general AI systems.
  </Card>

  <Card title="Shared Reference" icon="users">
    A common information source that all Space members can benefit from through the AI assistant.
  </Card>
</CardGroup>

## Benefits of Uploaded Files Retrieval

<Accordion title="Domain-Specific Expertise" defaultOpen={true}>
  Uploaded Files retrieval enables the AI assistant to develop specialized expertise in your unique domain. By accessing your organization's documents, the AI can:

  * Understand your specific terminology, products, services, and processes
  * Reference internal guidelines, policies, and best practices
  * Provide answers that reflect your organization's unique approach and perspectives
  * Deliver insights based on company-specific data and reports
</Accordion>

<Accordion title="Information Access and Consistency">
  The Uploaded Files ensures everyone gets consistent, accurate information:

  * Centralized access to important documents through <Glossary term="natural-language">natural language</Glossary> questions
  * Consistent answers across the organization based on the same source materials
  * Quick <Glossary term="retrieval">retrieval</Glossary> of information that might otherwise be buried in folders or drives
  * Reduction in time spent searching for information across multiple sources
</Accordion>

<Accordion title="Secure Information Utilization">
  Uploaded Files retrieval offers secure access to sensitive information:

  * Proprietary documents remain within your secure environment
  * The <Glossary term="privacy-filter">Privacy Filter</Glossary> automatically protects sensitive information during processing
  * Information access respects Space membership permissions
  * Your confidential data becomes useful without being exposed externally
</Accordion>

## How Uploaded Files Retrieval Works

When you ask a question in a Space with Uploaded Files, the AI assistant intelligently processes your <Glossary term="query">query</Glossary> and leverages relevant documents to formulate its <Glossary term="response">response</Glossary>.

<Steps>
  <Step title="Query Analysis">
    When you ask a question, the AI assistant analyzes what you're asking to determine if Uploaded Files documents might contain relevant information.
  </Step>

  <Step title="Document Retrieval">
    If relevant, the system searches the Uploaded Files to find documents containing information that might help answer your question.
  </Step>

  <Step title="Privacy Protection">
    Before processing, the Privacy Filter automatically detects and masks any sensitive information in the retrieved documents. When the Privacy Filter is enabled, images are also restricted from being sent to AI models not hosted within the EU.
  </Step>

  <Step title="Information Processing">
    The AI assistant reads and understands the relevant sections from the documents, extracting the information needed to answer your question.
  </Step>

  <Step title="Response Generation">
    Using both its general knowledge and the specific information from your Uploaded Files, the AI creates a comprehensive, contextual response.
  </Step>
</Steps>

<Tip>
  Uploaded Files retrieval works best with clearly structured documents and specific questions. For optimal results, ask direct questions about information likely to be in your documents.
</Tip>

## How to Recognize Uploaded Files Responses

When your AI assistant is using your Uploaded Files to answer questions, the experience looks noticeably different from regular AI responses. Here's what to look for:

<CardGroup cols={2}>
  <Card title="Progress Indicator" icon="circle-notch">
    During retrieval, a progress indicator at the top of the assistant's answer area shows real-time status updates like "Understanding query" and "Retrieving documents from Uploaded Files."
  </Card>

  <Card title="Retrieval Completion" icon="circle-check">
    When retrieval completes, the progress indicator shows a filled icon with a message indicating how many documents were found, such as "Retrieved 3 documents from Uploaded Files."
  </Card>

  <Card title="Multiple Retrieval Attempts" icon="arrows-rotate">
    The assistant may perform multiple retrievals during a single response if it needs additional information or if initial results weren't sufficient, with each attempt showing in the progress indicator.
  </Card>

  <Card title="Source References" icon="quote-left">
    The AI response often includes specific references to source documents, helping you understand where information originated and providing transparency about knowledge sources.
  </Card>
</CardGroup>

## Asking Effective Uploaded Files Questions

To get the most from Uploaded Files retrieval, phrase your questions in ways that help the AI assistant find and use the right information.

<AccordionGroup>
  <Accordion title="Be Specific and Direct">
    * **Good:** "What is our company's policy on remote work according to the employee handbook?"
    * **Better than:** "Tell me about working remotely"

    Specific questions help the AI pinpoint exactly what information to look for in your Uploaded Files.
  </Accordion>

  <Accordion title="Reference Document Types">
    * **Good:** "Based on our quarterly reports, how has our market share changed in 2023?"
    * **Better than:** "How is our market share doing?"

    Mentioning the type of document can help the AI narrow its search to the most relevant files.
  </Accordion>

  <Accordion title="Ask for Summaries and Analysis">
    * **Good:** "Summarize the key findings from our recent customer satisfaction survey"
    * **Good:** "What are the three main concerns raised in our project risk assessment documents?"

    The AI assistant can digest large documents and extract the most important points.
  </Accordion>

  <Accordion title="Request Comparisons">
    * **Good:** "Compare our current marketing strategy with the approach outlined in last year's plan"
    * **Good:** "What's different between our 2022 and 2023 compliance requirements?"

    The AI can identify similarities and differences across documents in your Uploaded Files.
  </Accordion>
</AccordionGroup>

## Common Questions

<AccordionGroup>
  <Accordion title="Does every Space need Uploaded Files?">
    No, Uploaded Files are optional. Spaces can function effectively without Uploaded Files for general topic-focused discussions. A Space with well-crafted Space Instructions but no Uploaded Files still provides a specialized AI assistant, just without access to your organization's documents.
  </Accordion>

  <Accordion title="Can the AI assistant understand all file types in the Uploaded Files?">
    The AI assistant can process common document formats including PDFs, Word documents, PowerPoint presentations, OpenDocument text/presentation formats, plain text files (TXT, JSON, Markdown, CSV, SVG), and images (PNG, JPEG, WebP, HEIC/HEIF, AVIF, BMP, GIF, ICO, SVG, and TIFF). The system extracts and indexes the text content from documents and analyzes visual content from images to make them searchable.
  </Accordion>

  <Accordion title="How recent is the information from my Uploaded Files?">
    The AI assistant accesses the current version of documents in your Uploaded Files at the time of your query. When you update documents in the Uploaded Files, the AI will use the new information in subsequent conversations.
  </Accordion>

  <Accordion title="Can I ask the AI to prioritize Uploaded Files information?">
    Yes. The most effective way is to use the **Simple Lookup** [Response Mode](/en/platform/core-features/chats/response-modes), which forces the AI to answer *only* using your Uploaded Files documents.

    You can also use **Auto** mode and instruct the AI by phrasing your question like: "Based only on our Uploaded Files documents, what is our policy on..."
  </Accordion>
</AccordionGroup>

## Best Practices

<CardGroup cols={2}>
  <Card title="Curate Your Uploaded Files" icon="hand-sparkles">
    Keep your Uploaded Files focused and relevant. Include only documents directly related to the Space's purpose to improve AI retrieval accuracy.
  </Card>

  <Card title="Update Documents Regularly" icon="arrows-rotate">
    Maintain current versions of documents in your Uploaded Files to ensure the AI provides the most up-to-date information available.
  </Card>

  <Card title="Use Clear Document Naming" icon="tag">
    Name your Uploaded Files documents clearly and descriptively to help with organization and make references in AI responses more meaningful.
  </Card>

  <Card title="Test Different Question Types" icon="flask">
    Experiment with various question formats to discover what works best for retrieving specific types of information from your Uploaded Files.
  </Card>
</CardGroup>

## Adding Uploaded Files to a Space

To start using Uploaded Files retrieval in your work, you'll need a Space with documents added to its Uploaded Files. Here's what to know:

* **Use Existing Spaces:** The easiest way to benefit from Uploaded Files retrieval is to use Spaces that colleagues in your organization have already created and shared with you
* **Join Team Spaces:** Many organizations create specialized Spaces with rich Uploaded Files for different departments or projects
* **Admin Permissions Required:** If you want to build your own Uploaded Files, note that only Space admins can add or manage documents
* **Your Own Space:** When you create a Space, you automatically become its admin
* **Simple Document Upload:** Adding documents is as simple as uploading files through the Space Settings

For a more detailed look at how to add and manage documents in a Space's Uploaded Files, see our page on [Managing Uploaded Files](/en/platform/core-features/spaces/managing-uploaded-files).

## Next Steps

Now that you understand Uploaded Files retrieval, you might want to explore:

* Learn how to [Create](/en/platform/core-features/spaces/creating-spaces) or [Customize a Space](/en/platform/core-features/spaces/customizing-spaces)
* Enhance collaboration by [Sharing Spaces with Colleagues](/en/platform/core-features/spaces/sharing-spaces) who need access to the same knowledge