0
8.0.6
Australia Patch 6, Australia Patch 5, Australia Patch 4, Australia Patch 3, Australia Patch 1, Australia, Zurich Patch 12, Zurich Patch 11, Zurich Patch 10, Zurich Patch 8, Zurich Patch 7, Zurich Patch 4, Zurich Patch 2, Zurich Patch 1, Zurich, Yokohama Patch 11, Yokohama Patch 8, Yokohama Patch 6, Yokohama Patch 3, Yokohama
Generative AI
The Group and Action Framework provides skills that group datasets into named clusters and then take action on the those clusters through generative AI to provide suggested steps, summaries, and recommendations.
Grouping skills: Classifies and clusters datasets stored in a table, generates human readable topics for each cluster, and selects samples to represent each cluster as examples within LLM prompts.
Action skills: Allows for LLM actions (such as generating resolution steps) on new records as they are created and on entire clusters.
--- AI Generated Release Notes ---
New
- Users can now classify items within GAF using Group Capability (Topic), enabling enhanced organization and management of records. System administrators can manage classification settings related to Group Capability (Topic).
- All third-party models are now supported for GAF KB Gap HR skills, expanding model selection options for skill processing.
- Taxonomy fields have been added and populated in the record group table for the KB GAP HR pipeline, allowing for improved categorization and searchability.
- Record groups are now replicated in action strategies for the KB GAP HR pipeline, ensuring consistent grouping across workflow actions.
- Topic generation can now run as the same user as the scheduled job, aligning execution context for automated topic creation.
Changed
- Incremental clustering for Agent Advisor now includes a refinement handler, improving noise clustering and prediction accuracy. The workflow has been updated to support filtering of the highest group in the group record table with an optional configuration flag.
- Incremental clustering implementation for Agent Advisor has been enhanced with noise clustering logic, input query filtering, and threshold adjustments for improved data grouping and workflow performance.
Fixed
- The HR GAF Group job now triggers correctly when the 'snhrcore.admin' role is removed from 'admin', ensuring group processing reliability.
- Field-level report view access control has been restored, allowing users with the data report viewer role to access additional data tables in dashboard visualizations.
- Skill Run Info for Agent Advisor has been corrected, resolving issues in the skillruninfo table query and ensuring accurate skill run data.
Not applicable for this application version.
- AI Search (Optional) - Required for the Optimized Prediction feature. This feature increases the maximum number of records to 500K, allows for faster online prediction, and increases quality of prediction/mapper results.