Internal Q&A Features

Directly engage in Q&A with AI Assistants on the platform. Related conversations are not displayed on external devices. Suitable for internal data processing, educational training, and other purposes.

Introduction to Internal Q&A

💬 Conversation Block

• Real-time Q&A interaction with AI Assistants

📋 Historical Conversation Records

• View and manage all past conversation records

🤖 AI Assistant Selection

• Switch between different AI Assistants for conversation

⚙️ AI Assistant Settings

• Real-time adjustment of current conversation AI Assistant settings, including permissions and tool configuration

📚 Knowledge Base Document Management

• Precisely control the scope of documents that AI Assistants can reference in each conversation

How to Use Internal Q&A?

1. Access the Internal Q&A Page

2. Select the AI Assistant You Want to Converse With

You can select the AI Assistant you want to converse with from the blue dropdown menu in the upper left corner.

3. Start Conversation

• Continue Past Conversation: Select the topic you want to continue from the past Q&A topics on the left side.

• Start New Conversation: Click "New Q&A" in the left section to start a completely new conversation.

All conversations will appear in the Q&A records in the upper left corner

After selecting a conversation, you can enter your question in the text box below to start the conversation.

Continue Past Conversation (Select Conversation)

Start New Conversation (New Conversation)

4. Customize Document Reference Scope (Optional)

Use the knowledge base management function on the right side to select specific documents for the AI Assistant to reference in this conversation.

• Each message can freely select reference documents and FAQs

• Can be reconfigured before sending each Q&A

Document Access Scope

1. All documents are available when the first message is sent, including "Ambient Light-Series.md"

AI Assistant responds after referencing: "Document content: The official website has been closed and no longer sells any products to the public"

1. Uncheck "Ambient Light-Series.md" when sending the second message

AI Assistant cannot reference it and responds: "I cannot find the specific content of the 'Ambient Light-Series.md' file."

FAQs Access Scope

1. All FAQs are available when the first message is sent, including "What is the operating principle of LLM?"

AI Assistant responds with relevant answer after referencing

1. Uncheck "What is the operating principle of LLM?" when sending the second message

AI Assistant cannot reference "What is the operating principle of LLM?" and responds "I cannot find information about LLM operating principles in the FAQ"

Tag Filtering

You can use the tags set in the knowledge base to filter the document content you want to make available to the AI Assistant:

For tag settings, please refer to: Document Management: Tags and Metadata

a. Click "Show Tag Filter" in the upper right corner

b. Select AND or OR filter conditions, click the "AND" text to switch

c. Add tags

• Using "AND" filter

  • Documents: Only "Advanced Camping.pdf" meets all conditions and can be referenced by AI Assistant

  • FAQ: No FAQ meets all conditions simultaneously, so none are displayed

Tag filtering will be applied to both documents and FAQ content simultaneously

5. Real-time AI Assistant Settings Adjustment (Optional)

Through "AI Assistant Settings" in the upper right, you can modify permissions, tools, and other configurations in real-time.

After clicking, a right-side popup window will appear, allowing you to freely edit the required permissions, tools, etc.

💡 Usage Tips

Conversation Management

• Make good use of historical conversation records to quickly return to previous discussions

• New Q&A is suitable for starting completely new topics

Document Control Precision

• For specific questions, limit reference documents to get more precise answers

• Avoid irrelevant documents interfering with the AI Assistant's judgment

Flexible Settings Adjustment

• Adjust AI Assistant capabilities in real-time for different conversation scenarios

• Test the impact of different settings on answer quality