Multi-agent llm narrative generation architecture

A computer-implemented method, comprising:

• generating, by a generation agent executed by the computing system, an event narrative draft based at least in part on case data associated with alerted interactions from a monitoring system, wherein the case data includes interaction records, alert highlights, and entity information;
• parsing, by a pattern detector agent executed by the computing system, the event narrative draft to automatically identify one or more candidate statements that aggregate or generalize the case data in the event narrative draft;
• distributing, by a master evaluator agent executed by the computing system, the one or more candidate statements to one or more pattern checking agents that correspond to the one or more patterns for evaluation for correctness;
• executing, by the one or more pattern checking agents executed by the computing system, a programmatic validation of individual candidate statements against the interaction records to classify the individual candidate statements as consistent with or inconsistent with the interaction records;
• and
• combining, by the master evaluator agent executed by the computing system, the validated statements from the one or more pattern checking agents into a final event narrative based on the corrected statements.

A computing system, comprising:

• one or more computing devices operably connected to communicate over one or more communication networks via one or more network interfaces;
• at least one processor connected to at least one memory, wherein the at least one processor is operably connected to at least one of the one or more computing devices;
• and
• one or more non-transitory computer readable media including instructions stored thereon that when executed by at least the computing system cause the processor to:
• generate, by a generation agent executed by the computing system, an event narrative draft based at least in part on case data associated with alerted interactions from a monitoring system, wherein the case data includes interaction records, alert highlights, and entity information;
• parse, by a pattern detector agent executed by the computing system, the event narrative draft to automatically identify one or more candidate statements that aggregate or generalize the case data in the event narrative draft;
• distribute, by a master evaluator agent executed by the computing system, the one or more candidate statements to one or more pattern checking agents that correspond to the one or more patterns for evaluation for correctness;
• execute, by the one or more pattern checking agents executed by the computing system, a programmatic comparison of individual candidate statements against the interaction records to classify the individual candidate statements as consistent with or inconsistent with the interaction records;
• and
• combining, by the master evaluator agent executed by the computing system, the validated statements from the one or more pattern checking agents into a final event narrative based on the corrected statements.