Release Notes v. 7.1.0.0

June 2025 - SmartHub 7.1.0.0

This page outlines all of the updates and enhancements in this release of SmartHub. For more information on upgrading, see Upgrading SmartHub.

The following features and enhancements have been added to SmartHub in this release:

System enhancements

Feature Description
Related documents in split-screen view

This release of SmartHub adds support for displaying related documents in a split-screen view, allowing users to receive the functionality of traditional SmartHub search and SmartHub conversational search from a single page.

When enabled, a new pane displays with related results that are most similar to the user's last conversational search message. For more information, see Configure the conversational search module for your SmartHub pages.
Support for conversational search filters

You can now configure search refinement filters that users can apply to their Conversational Search chats to fine-tune the AI-generated responses for parameters such as author, category, format, etc.

These filters work similar to SmartHub refiners, but they are applied to the knowledge sources that are used by your chosen LLM to provide answers. After applying filters, they will be taken into consideration when a subsequent question is asked. They will not be used for any question that was asked before the filter was applied. Users also have the option to clear their applied filters, and all filters will be cleared when a new conversation is started or the page is refreshed.

For more information, see Using Conversational Search filters.
Document previews in chat interface

When using SmartHub conversational search, users will now be able to preview the relevant sources that powered the conversational search bot's response. By clicking the Sources button under the response, a list of documents will display. Selecting one of these documents will allow users to go directly to the source or preview the document.

When previewing the document, a new panel will open to the right of the conversational search response, opening the preview on the relevant page from the document. For more information on configuring Smart Previews settings for conversational search, see Configure the conversational bot settings.
RESTful query tuning stage

This release of SmartHub adds the RestfulPostman tuning stage, allowing you to use the functionality of any Restful API (LLMs, AI Cloud Services, etc.) of your choosing to enhance your user's search experience.

For more information, see Apply Restful API functionality to your search queries.
Vector embedding generator tuning stages

This release of SmartHub adds two new user experience tuning stages for Azure OpenAI and OpenAI vector embedding generation. These stages can automatically transform user search queries to apply vector embeddings to the query during runtime.

For more information, see Apply vector embeddings to user queries with the OpenAI embeddings generator and Apply vector embeddings to user queries with the Azure OpenAI embeddings generator.
SmartHub Analytics enhancements

The SmartHub analytics mechanism has been updated with the following enhancements:

  • SmartHub analytics now provides detailed information on the state of a user's search query. This includes the keywords, as well as all refiner properties and values that were applied to the query when it was executed. For more information, see Events captured in the SmartHub Analytics default schema.

  • The search module now tracks when a refiner is removed, recording information about the query, the refiner, and the user. For more information, see Tracking user queries and selections
Elasticsearch Authentication enhancements

You can now configure your Elasticsearch search engine to use an API key for authentication.

For more information, see Configure your Elasticsearch search engine to use API key for authentication.
Support for Vector search with Elasticsearch and OpenSearch Elasticsearch and OpenSearch search engines now support vector search and hybrid search queries. In the Elastic Translator user experience tuning stage, you can enable the Enable Pure Vector Search field to allow SmartHub to submit pure vector or hybrid search queries. For more information, see Translate queries into an Elastic search query with the Elastic Translator
Updated behavior for pure vector searches with Azure AI Search

Pure vector search is now enabled by default when using the Azure Query Translator user experience tuning stage. If a user's search query does not have vector embeddings, this setting is ignored.

For more information, see Translate queries into an Azure search query with the Azure Query Translator.
Date Refiner support for Microsoft Search Items

In this release of SmartHub, enhancements were made to flatten specific properties received from the Microsoft Graph API. Depending on the Microsoft Graph response structure, this update may remove the need to map to the tree-based metadata structure from the Microsoft Graph API response. When configuring the property mapper tuning stage, you now only need to provide the property names for driveItem and listItem fields, and externalItem properties for your mapping.

If you are upgrading from a previous version of SmartHub, the property mapper tuning stage must be reconfigured.

For more information, see Microsoft Search Items.
Conversational search enhancements

You can now specify the text rendering mode for your conversational bot's response. The following 3 modes are available:

  • character: This instructs the page to render each letter of a received chunk, one by one.

  • chunk: This instructs the page to render each chunk of the bot response, one by one.

  • nonstream: This instructs the page to render the full bot response all at once. This will use the FullMessage event instead of the Message event.

For more information, see Configure the conversational search module for your SmartHub pages.

Fixed issues

The following issues have been fixed in this release:

  • When SmartHub 7.x was configured with an on-prem SOLR instance, no search results were returned. (BAI-18144)

  • When multiple Microsoft Search API search engines were mixed, the ranked order of the search results was reversed. (BAI-18082)

  • If you redirected your search to SmartHub from a SharePoint Online page with SmartHub type-ahead integration with the SavedQueries provider enabled, the query would redirect to an invalid SharePoint page. (BAI-18072)

  • If your SmartHub results page was configured to return results from multiple Microsoft Search API search engines, any subsequent pages of results would return the same results as the first page. (BAI-18067)

  • Applying a refiner to search results would reset any sorting applied. (BAI-18047)

  • On a SharePoint Online page with SmartHub type-ahead integration, formatting and display issues were present. (BAI-18012

  • When configuring a vertical and changing the sorting, SmartHub would incorrectly use the default template for the vertical. (BAI-17996)

  • Certain metadata properties were not being returned for Microsoft Search API search engines. (BAI-17994)

  • When using Conversational search in hybrid mode, multiple calls to the API were made, resulting in a mixed response from the bot. (BAI-17986/BAI-18020)

  • In some scenarios, XRANK boosting was being applied to results incorrectly. (BAI-17865)

  • When a search engine was mixed with any Microsoft Search API search engine, a "Specified cast is invalid" error was returned. (BAI-17860)

  • If Expertise search contained the ContentSource refiner, an error was returned. (BAI-17842)

  • In some Scenarios, the analytics reports site was not showing reports. (BAI-17807)

  • In some scenarios, when SmartHub was integrated with Microsoft Dynamics, the search bar did not load on the landing page. (BAI-17789)

  • In some scenarios, the response from the Conversational Search module was displayed incorrectly. (BAI-17661)

  • When filtering the Expertise page by Department, the filter was not refining the job list. (BAI-17646)

BREAKING CHANGES

The following are breaking changes that may affect your integration. Please ensure that you carefully review this list before upgrading to SmartHub 7.0 and later:

  • When upgrading to SmartHub 7.0 and later, you must update any custom pages to preserve your existing CSS or upgrade to the out-of-the-box SmartHub 7.0 page layouts. For more information on this update, see Update or preserve your custom CSS in the upgrade documentation.

  • The DomContentLoaded event has been removed in SH 6.2. You will need to replace the event with the afterInitialization event. For more information, see UI events.