Frequently asked questions

Assist is designed to respond to prompts related to the documents in the current workspace and not general knowledge prompts. Write explicit and concise prompts using natural language, such as "Who are the key people in this case?" or "Which documents mention [topic]?" For more guidance on writing prompts, see Best practices.

There is no limit to the number of prompts allowed, but long or highly complex queries may take more time to process. For best results, keep your prompts concise and focused. For more guidance on writing prompts, see Best practices.

Yes. Assist uses advanced intention capabilities to reject prompts that cannot be answered or are outside its scope. It also requests clarification for vague or overly broad queries. See Intent detection for more information.

Assist uses a Retrieval-Augmented Generation (RAG) approach to respond to prompts. The system first retrieves the most relevant documents for the query and then uses a large language model (LLM) to generate a summarized response. The following best practices can help improve accuracy when working with contract, entity, or relationship extraction tasks. For the full list, see Best practices:

• Be explicit, not implicit
  Retrieval systems work best with clearly defined prompts. Vague prompts, such as “Tell me about the case,” are less effective than specific ones, like “Describe contract terms related to indemnification.”
• Write concise prompts
  Multi-part or compound prompts can fragment retrieval results. Breaking complex prompts into smaller, focused ones improves precision and clarity.
• Leverage keywords and synonyms
  Retrieval engines benefit from varied phrasing. Include alternative terms or related entities, such as “bribe,” “gift,” or “incentive,” to capture a broader range of relevant content.
• Lack of comprehensiveness
  Retrieval systems aim to respond to the prompt directly, rather than provide a full list of all related documents or entities.

The large language model (LLM) used by Assist has been evaluated for 83 languages. Although Assist has primarily been tested on English-language documents, it is designed to support non-English datasets. For more information on this issue, see Language support.

Individual prompts themselves cannot be shared with others, however, you can share conversations (prompts and responses) with others in the workspace if you have the proper permissions to do so. See Sharing a conversation and Permissions for details.

You must set up their permissions to only allow access to Prompting but not to Index Management. Similarly, if you have users who are only responsible for managing indexes (create, rebuild, delete), you could set them up with only Index Management permission, but not Prompting. See Permissions for details.

Yes. You click the star icon below a response. See Starring (bookmarking) a prompt and response for details.

Yes. Important conversations threads can be pinned to the top of the Conversations list. They will appear under the Pinned section. See Pinning a conversation for details.

No. Assist does not currently include configuration options to control response variability or “thinking time.” The system used by Assist is designed to reduce variability, but some level of randomness is inherent to all large language models (LLM). As a result, identical prompts may generate slightly different responses, especially when phrased implicitly or influenced by prior conversation context. The more specific and well-defined the prompt, the more consistent the output tends to be. Assist focuses on accuracy and relevance rather than adjustable response styles. For more information, see Assist (aiR Assist).

Not at this time. Assist follows a Retrieval-Augmented Generation (RAG) approach, in which only the most relevant document chunks are retrieved and used to generate a response. Repeated or boilerplate text (such as, standard disclaimers, email footers, or signatures) is not explicitly filtered out. However, this type of content is typically not selected unless it is determined to be highly relevant to the user’s query.

Assist relies on indexed data to generate responses leveraging retrieval-augmented generation (RAG). While it aims to offer correct responses, always verify the offered information and citations for accuracy. Using the feedback buttons in the response box can help improve the tool over time. See Feedback options for details on using the feedback buttons.

To leave feedback in the Assist app, use the thumbs up or down icons below each response to rate it and add comments. See Feedback options for more information. You may also occasionally be present with an Assist Feedback dialog asking how helpful were Assist's responses for completing your task. See In-product feedback survey for details.

Each user's session and conversation history remains private, however, a user with the proper permissions can share specific conversations with others in the workspace. See Sharing a conversation and Permissions for details.

Yes. You can delete an entire conversation and its history using Conversation Manager. See Using Conversation Manager for details.

Your conversation history is saved across your user sessions and retained as long as the workspace exists, unless the conversation is deleted. See Using Conversation Manager for details on deleting conversations. Also refer to Conversation history and sessions for more information surrounding conversation history.

Yes. Provided you have the proper permissions, you can share conversations with others in the workspace. See Sharing a conversation and Permissions for details.

Yes. You can rename an entire conversation using Conversation Manager. See Using Conversation Manager for details.

Yes. You can switch between conversations anytime, as well as return to a previous conversation to continue querying using Conversation Manager. See Using Conversation Manager for details.

Yes. Multiple users can use Assist within the same workspace concurrently. Each user's session and chat history remain private and separate, although users can choose to share particular conversations with others in the workspace. See Sharing a conversation for more information.

Yes. Assist maintains conversational continuity within a session, allowing it to reference the context of previous queries when formulating responses. This means that follow-up prompts can build on information discussed earlier in the same conversation.

However, the continuity mechanism in Assist differs from general-purpose chat systems, such as ChatGPT. Each interaction still begins with a new document retrieval step, where Assist searches for the most relevant materials based on the current query and prior context. The retrieved content then informs the contextualized prompt, which the language model uses to generate the final response.

Yes. Provided you have the proper permissions, conversations can be exported to an MS Word (.docx) file without losing the context, citations, or source links. See Exporting a conversation and Permissions for details.

Assist indexes do not currently have a dedicated view within Relativity aiR. Indexes are managed directly through the Assist interface. When switching between indexes, the system displays a status card within the chat showing the index state.

If an index was built with errors, the status card includes a “View Errors” option that lists documents that encountered issues during indexing. Documents larger than 5 MB of extracted text are automatically skipped during the indexing process and are not currently displayed as errored items.

See Working with indexes for more information.

You must set up their permissions to only allow access to Index Management. See Permissions for details.

Assist generates its responses based solely on the extracted text from processed documents. This means that for Excel files, the content available to Assist depends on how text is extracted during processing.

Depending on your processing profile settings and the engine used to extract text, certain elements (such as data from multiple tabs, formulas, or embedded objects) may or may not be included in the extracted text. To ensure the most complete and accurate results, review and optimize your processing configuration before indexing Excel files.

Assist can only use the extracted text available in Relativity aiR. For image-only documents (such as scanned PDFs or pictures without embedded text), text extraction must be performed in advance, either through an external process or using an OCR (Optical Character Recognition) workflow within Relativity aiR.

Without extracted text, Assist will not be able to index or generate responses from image-based documents.

No additional steps are required for large documents that are within the 5 MB of extracted text size limit. During indexing, Assist automatically divides large documents into smaller text chunks to optimize retrieval and improve response relevance. These chunks may be retrieved individually when Assist processes a query, ensuring that large documents are handled efficiently without additional configuration.

No. Documents exceeding the 5 MB of extracted text size limit are automatically skipped during indexing and are not currently flagged or listed in the error report. As a result, these files will not appear among errored documents in the Assist interface.