Sign in

Week of September 11th

New updates include: Conversation Assist & Knowledge AI, Conversational Cloud Infrastructure and Salesforce Connector

Exact delivery dates may vary, and brands may therefore not have immediate access to all features on the date of publication. Please contact your LivePerson account team for the exact dates on which you will have access to the features.

The timing and scope of these features or functionalities remain at the sole discretion of LivePerson and are subject to change.

Conversational Cloud Infrastructure

Features

Generative AI - More ways to manage your use of LLMs via LivePerson

If you’re using an LLM made available via LivePerson to power your Generative AI solution in Conversational Cloud, this note is for you.

In this release, we introduce an LLM Management page to the Management Console. Use this new page to manage your use of the LLMs made available via LivePerson:

  • Expose specific models for use in prompts: For various reasons (performance, cost, etc.), you might want to expose only specific LLMs to your prompt engineers who are creating the prompts for your Generative AI solution. And you can. Only the models that you select are exposed in the Prompt Library that’s used to create and manage prompts.
  • View a model’s RPM limit: Want more visibility into limits, so you can use this data to inform your decisions about model selections? It’s here. The Requests Per Minute limit of each model is now visible in the UI.
  • Change the default models: You can customize which LLMs are used by default in prompts in the Prompt Library. Prompt engineers remain free to select a different model (from the list of models that you’ve specified as allowed).
Image

Generative AI - Onboard and manage your use of your in-house LLM

If you’ve invested in an in-house LLM, you can use it to power the Generative AI features in your Conversational Cloud solution. This lets you align the solution with your brand’s overall LLM strategy.

Image

LivePerson now offers a self-service UI for onboarding and managing your use of your in-house LLM in your Conversational AI solution. Use the new LLM Management page to the Management Console to:

  • Add an LLM subscription: Add as many subscriptions as you require.
Image

No credentials beyond what the LLM provider requires are needed.

  • Change the default models: You can customize which LLMs are used by default in prompts in the Prompt Library. Prompt engineers remain free to select a different model (from the list of models that you’ve specified as allowed).
Image

  • Disable or enable a subscription: As you’re testing models, you might want to hide certain models from users who are creating prompts in LivePerson’s Prompt Library. Disabling a subscription hides the subscription (and, by extension, its associated models) from users, so they can’t select it for use in a prompt. You can disable and re-enable a subscription at any time.
  • Delete a subscription: Delete a subscription that you no longer have use for. Once the subscription is deleted, prompts that use the associated models are automatically switched to use a default model. This ensures the prompts continue to work.

Conversation Assist

Features

Offer answer recommendations in voice conversations

It’s here! You can now use Conversation Assist to offer answer (article) recommendations to agents in voice conversations.

Image

Image

The following features are supported in voice conversations:

  • Conversation Assist can offer answer recommendations to the agent based on the intent in the consumer’s utterance. The recommendations are based on the knowledge base-level configuration rules that you set up. (Bot recommendations aren’t supported.)
  • Agents can also look up answers on demand.
  • Agents can mark recommendations as used (click the Use button), so this info appears in reporting. This info helps you to understand exactly who’s using what. The existing reports have been updated to indicate whether the recommendation occurred in the context of a messaging or voice conversation.
  • Agents can provide “thumbs up” and “thumbs down” feedback on recommendations. This info helps you to understand which answers need optimization.

Setup: Once voice is enabled on your LivePerson account, use the Manage > Users & Skills area to set up your voice agents. We recommend that you assign only one skill to your agents who handle voice conversations. In practice, Conversation Assist uses the first skill that’s assigned to the agent to determine whether to offer recommendations. But, if there’s just one skill assigned, there’s no question as to which skill is being used.

Image

KnowledgeAI

Features

New and improved LivePerson prompt for answer enrichment via Generative AI

If you’re using Generative AI to enrich the answers from your knowledge base before:

We have great news on this front! 

LivePerson’s research team is always hard at work developing and testing prompt improvements, and this commitment and vigilance has paid off. We are pleased to announce that there’s a new LivePerson prompt available for use in messaging solutions. The new prompt enhances the accuracy of responses from the LLM by reducing hallucinations. When compared to our previous prompt for answer enrichment, this new prompt performs 3x better!

We highly recommend that you switch to this new prompt for more reliable and contextually accurate answers from KnowledgeAI.

What’s the name of the new prompt?

  • Conversation Assist’s new prompt is “Enrichment Factual EN - Conversation Assist v2.”
  • Conversation Builder’s new prompt is “Enrichment Factual EN - Messaging bot v2.”

Is this new prompt available as a LivePerson template?

Yes, it is. Learn how to create a prompt by copying a LivePerson template.

Does this change impact my Generative AI solution?

No, it doesn’t. We are simply making the new prompts available as LivePerson prompt templates.

Is there a call to action for me?

Yes, since this prompt performs 3x better than the old prompt, we highly recommend that you test this prompt and integrate it into your solution as soon as you can.

Important note

We’re so pleased with the performance of these new prompts that soon we’ll be changing the default behavior in Conversational Cloud. 

On September 23, 2024, LivePerson will change the system so that these new prompts are used by default moving forward. This means that after September 23:

  • If you add a new knowledge base-level rule in Conversation Assist, it will use the new prompt.
  • If you add a new KnowledgeAI interaction in a bot in Conversation Builder, it will use the new prompt.

Salesforce Connector

Enhancements

Salesforce Connector

LivePerson are pleased to announce feature enhancements to the Salesforce Connector including:

Support for Custom Away States- Support for custom away states have been added to the Salesforce Connector frontend. Custom away states may be configured in the “Manager Workspace” in CCUI.

Enhanced Logout Experience - Agents will be logged out from the Salesforce Connector within 15 seconds if the same agent logs into CCUI or from another instance of the Salesforce connector. A feedback message will be presented to tell the agent why they were logged out.

Enhanced Search Experience  - New endpoints have been implemented to search by resource ID (such as Case ID). To use this in the app, configure an advanced workflow trigger with the {{relateInteractionFromEngagementAttr}} action to automatically relate a case or opportunity to an interaction. After relating the resource the app will attempt to automatically match the interaction to the record (Contact/Account/Lead) associated with that resource (Case or Opportunity) in Salesforce.

UI configuration options for match modal and notes/disposition components - New configuration options have been added to the user interface for the match modal and notes/disposition components. These changes allow admins or users to customize the appearance and behavior of the interface where agents match records or cases with relevant customer data & notes / outcomes of customer interactions.

Enhanced Dispositions - Added support for multiple layers of field value dependencies improving flexibility and precision when categorizing customer interactions, enabling more structured and detailed data collection.

Improved performance of NLP parsing for note input - the system now processes and understands the text agents enter into notes more efficiently using Natural Language Processing (NLP). This improvement enhances the speed and accuracy of analyzing and interpreting agent notes

Changed UX to permit sending an attachment without requiring text input - Agents can now send attachments to customers without the need to include a message or text.Previously, agents were required to type a message alongside an attachment, but with this update, they can simply send a file on its own, streamlining the process and making it more efficient in scenarios where no additional explanation is necessary.

Performance Improvement - Performance Improvement enhances system efficiency, speed, and reliability, resulting in faster data processing, reduced latency, and a smoother experience for agents and customers.

Bug Fixes:

Fixed edge case causing rich text formatting icons to not show sometimes for the Web messaging and Email channels.