Asking a question

You can ask aiR Assist questions about indexed documents using conversational language. Asking follow-up questions builds on the previous one. aiR Assist follows the context of the conversation and people or events discussed. Refer to Best Practices for tips on writing questions.

Navigating the Question box and Responses

To access aiR Assist, click its icon () in the left nav bar.

Navigating the Ask a question box

Refer to the following table to learn how to use the Ask a question box.

Number	Name	Description
1	Ask a question	Type your question into this box. Questions must be at least three characters long to submit.
2	Index used	Name of the index used for the query. Hover over the name to view the number of documents within the index, date/time last indexed and by whom, and description entered when the index was created.
3	Start new chat conversation	Click to start a new chat conversation below the previous one, separated by a line. The new chat conversation does not reference previous conversations for its responses.
4	Send message	After typing your question, click the up arrow to submit it for a response.

Navigating the responses

After submitting a question, you will receive responses and additional information, as described below. For more information on responses, see Understanding aiR Assist responses.

Number	Name	Description
1	Question Asked	The question that was asked, along with the user's initials.
2	Natural Language Response	The answer to the question based on the retrieved documents, with inline citations to specific References.
3	List of References	Click to show or hide the full list of referenced documents generated for the response. aiR Assist can return up to 25 references in response to a single question.
4	Document Control Number	In the list of references displayed, the first few sentences of each document shows. Click More to view additional content if it exists. Click on a document control number link in the reference list to open the document in the Viewer.
5	View in the Document List	Click to filter the Document List to only show documents that include a reference. The selection of a specific folder in the Folder Browser may affect the visibility of references in the Document List. To ensure all references are visible, select the root folder level.
6	Index Name	Name of the index used for the query. Hover over the name to view the number of documents within the index, date/time last indexed and by whom, and description entered when the index was created.
7	Action buttons	Click the action buttons to do the following: Copy icon () —use to copy the entire response, including citations, displayed to the clipboard. See Copy answer for more information. Thumbs up icon ()—use to indicate positive reaction to the answer provided. See Feedback controls (thumbs up/thumbs down) for more information. Thumbs down icon ()—use to indicate negative reaction to the answer provided. See Feedback controls (thumbs up/thumbs down) for more information.

Copy answer

The Copy action lets you copy the full response, including citations, to your clipboard with formatting intact for use elsewhere. No need to manually transcribe the content.

Feedback controls (thumbs up/thumbs down)

The feedback buttons allow evaluation of the quality of the generated response:

• Thumbs Up —indicates the response was accurate, useful, or aligned with expectations.
• Thumbs Down —indicates the response was incomplete, incorrect, unclear, or not aligned with expectations.

Feedback submitted through these controls is collected in the background and used to:

• Monitor overall answer quality
• Identify recurring issues
• Improve future model behavior and tuning

No additional prompts or follow-up questions are displayed after submitting feedback.

Asking a question

To ask a question:

1. Click the aiR Assist () icon in the left nav bar.
2. Do one of the following:
   • To use the current index displayed in the lower left corner of the Ask a Question box, go to the next step.
   • To use a different index to query, click the index name displayed in the lower left corner of the Ask a Question box, select another index from the Index Manager list, then click Set index.
3. Type your question in the Ask a question box using natural language, as if chatting with a person.
   
4. Click the up arrow () to submit it. aiR Assist displays a response that includes a list of references and documents with control numbers. The application can return up to 25 references in response to a single question. We recommend reviewing the accuracy of the answer and its citations. For additional information on responses, see Navigating the responses and Understanding aiR Assist responses.
5. Continue asking questions using the selected index. If you want to use a different index for querying, click the current index name link, select another one from the index list, and click Set. Then, repeat these steps.

The system maintains chat context within the current sessions for follow-up questions and more natural conversation.

Chat history for all chat sessions:

• are private to each user.
• persists across each user's Relativity session.
• cannot be deleted.

Starting a new chat conversation

You can start a new conversation using the same or different index.

In the Ask a question box, simply click the New conversation () icon and begin typing your query in the box.

Each conversation session is separated by a line and does not refer to earlier conversations for its responses.

Conversation history and sessions

aiR Assist keeps a continuous conversation history for each user, ensuring consistency across questions in a session. Each user’s conversation data is kept private and is not available to others within the workspace. The conversation history is also saved between user sessions, letting users scroll back to review their earlier chats as needed.

It is important to note that this conversational memory for follow-up questions is limited by the context window size of 10 previous messages within the current user session. As the conversation grows longer, earlier parts of the exchange may no longer be referenced in responses once the model’s context capacity is reached.

Intent detection

Intent detection is one of the mechanisms that aiR Assist uses to understand the purpose behind each question and respond in the most appropriate way. Instead of simply matching keywords, aiR Assist analyzes the intent of a query to determine whether it should:

• Retrieve and summarize information from indexed documents.
• Ask for clarification when the question is unclear or refers to multiple topics.
• Indicate when a request cannot be fulfilled due to missing data or unsupported actions.

This capability ensures that aiR Assist provides relevant, accurate, and context-aware answers, resulting in a more natural and efficient chat experience.

How intent detection works

When a question is submitted, aiR Assist automatically evaluates its intent and decides how to respond. Depending on the context and available data, one of the following response strategies is used:

Response strategy	Description	Recommended actions
Retrieval-based answering (using RAG)	When a question can be answered using the indexed documents, aiR Assist retrieves the most relevant information and generates a concise, well-supported response.	No action needed.
Refusal due to missing data	If the system cannot locate relevant information within the indexed dataset, aiR Assist will clearly indicate that no matching content is available.	Verify that the appropriate index is selected. Confirm that the expected documents are included in the saved search used to build the index. Rephrase the question to be more specific (for example, include names, dates, or topics). If information truly does not exist in the indexed content, no further action is required.
Refusal due to unsupported request	When a question asks for something beyond aiR Assist’s scope - such as legal opinions, drafting documents, or making recommendations - the system identifies the intent and responds appropriately.	Rephrase the request as a document-based question (such as, “What documents discuss termination terms?” instead of “Should we terminate the contract?”). Focus on facts, references, or evidence rather than conclusions or advice.
Disambiguation (clarifying ambiguous queries)	If a question references multiple possible entities or events, aiR Assist will prompt for clarification before continuing.	Provide the clarification asked for.
Fallback for Unclear Intent	If the system cannot confidently determine what the question refers to, it will ask a clarifying question. If clarification is not provided, aiR Assist will default to generating a retrieval-based response using the available dataset.	Add relevant detail such as people, timeframe, or subject matter. Break broad questions into smaller, more focused inquiries. Confirm whether the goal is to find facts, summarize documents, or surface key themes.