US 12,437,213 B2
Bayesian graph-based retrieval-augmented generation with synthetic feedback loop (BG-RAG-SFL)
Andrew Denis, Prescott, AZ (US); and Harry Howard Wills, Jr., Princeville, HI (US)
Assigned to ZON GLOBAL IP INC., New York, NY (US)
Filed by Seer Global, Inc., New York, NY (US)
Filed on Nov. 1, 2024, as Appl. No. 18/935,013.
Application 18/935,013 is a continuation in part of application No. 18/787,514, filed on Jul. 29, 2024, granted, now 12,236,964.
Claims priority of provisional application 63/541,891, filed on Oct. 1, 2023.
Claims priority of provisional application 63/529,724, filed on Jul. 29, 2023.
Prior Publication US 2025/0217671 A1, Jul. 3, 2025
Int. Cl. G06F 40/30 (2020.01); G06N 5/022 (2023.01); G06N 5/043 (2023.01); G06N 7/01 (2023.01)
CPC G06N 5/022 (2013.01) [G06N 5/043 (2013.01); G06N 7/01 (2023.01)] 5 Claims
OG exemplary drawing
 
1. A method for enhancing the accuracy and relevance of generated responses, comprising:
receiving a user query;
performing graph-based retrieval from a knowledge graph;
at least one large language model (LLM) generating a response to the user query;
performing Bayesian evaluation of the response to the user query and determining whether the response meets a predetermined quality threshold;
performing secondary ground truth verification on the response;
verifying the response with multiple artificial intelligence (AI) agents;
wherein the multiple AI agents include a fact-checker, a coherence analyzer, a relevance assessor agent, and an ethical compliance agent;
adjusting the response based on feedback from the multiple AI agents; and
delivering the response to a user device.