https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/Head https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8 http://www.nanopub.org/nschema#hasAssertion https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/assertion https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8 http://www.nanopub.org/nschema#hasProvenance https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/provenance https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8 http://www.nanopub.org/nschema#hasPublicationInfo https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/pubinfo https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8 http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://www.nanopub.org/nschema#Nanopublication https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/assertion https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/dc/terms/title PersonalAI: A Systematic Comparison of Knowledge Graph Storage and Retrieval Approaches for Personalized LLM agents https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/describes https://neverblink.eu/ontologies/llm-kg/methods#aStarRetrievalAlgorithm https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/describes https://neverblink.eu/ontologies/llm-kg/methods#beamSearchRetrievalAlgorithm https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/describes https://neverblink.eu/ontologies/llm-kg/methods#memorizePipeline https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/describes https://neverblink.eu/ontologies/llm-kg/methods#mixedRetrievalAlgorithm https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/describes https://neverblink.eu/ontologies/llm-kg/methods#personalAiFramework https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/describes https://neverblink.eu/ontologies/llm-kg/methods#qaPipeline https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/describes https://neverblink.eu/ontologies/llm-kg/methods#waterCirclesRetrievalAlgorithm https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#ariGraph https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#graphRAG https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#hippoRAG https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#knowledgeGraphPrompting https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#memWalker https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#pearl https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#readAgent https://doi.org/10.48550/arXiv.2506.17001 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#retrievalAugmentedGeneration https://doi.org/10.48550/arXiv.2506.17001 http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://www.w3.org/ns/prov#Entity https://neverblink.eu/ontologies/llm-kg/methods#aStarRetrievalAlgorithm http://purl.org/dc/terms/subject https://neverblink.eu/ontologies/llm-kg/categories#SynergizedReasoning https://neverblink.eu/ontologies/llm-kg/methods#aStarRetrievalAlgorithm http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#aStarRetrievalAlgorithm http://www.w3.org/2000/01/rdf-schema#comment This method is a specific graph traversal algorithm, based on A*, designed for the QA pipeline to extract relevant triples from the memory graph. It employs various heuristics (Inner Product, Weighted Shortest Path, Averaged Weighted Shortest Path) to optimize pathfinding, contributing to the synergized reasoning capabilities of the overall framework. https://neverblink.eu/ontologies/llm-kg/methods#aStarRetrievalAlgorithm http://www.w3.org/2000/01/rdf-schema#label A* Retrieval Algorithm https://neverblink.eu/ontologies/llm-kg/methods#aStarRetrievalAlgorithm https://neverblink.eu/ontologies/llm-kg/hasTopCategory https://neverblink.eu/ontologies/llm-kg/top-categories#SynergizedLLMKG https://neverblink.eu/ontologies/llm-kg/methods#ariGraph http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#ariGraph http://www.w3.org/2000/01/rdf-schema#label AriGraph https://neverblink.eu/ontologies/llm-kg/methods#beamSearchRetrievalAlgorithm http://purl.org/dc/terms/subject https://neverblink.eu/ontologies/llm-kg/categories#SynergizedReasoning https://neverblink.eu/ontologies/llm-kg/methods#beamSearchRetrievalAlgorithm http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#beamSearchRetrievalAlgorithm http://www.w3.org/2000/01/rdf-schema#comment Inspired by LLM token generation, this retrieval algorithm constructs multiple semantically relevant paths from a starting vertex in the knowledge graph. It is a key component of the QA pipeline, governed by hyperparameters for depth, path count, and intersection rules, and contributes to the synergized reasoning process by finding optimal information retrieval paths. https://neverblink.eu/ontologies/llm-kg/methods#beamSearchRetrievalAlgorithm http://www.w3.org/2000/01/rdf-schema#label BeamSearch Retrieval Algorithm https://neverblink.eu/ontologies/llm-kg/methods#beamSearchRetrievalAlgorithm https://neverblink.eu/ontologies/llm-kg/hasTopCategory https://neverblink.eu/ontologies/llm-kg/top-categories#SynergizedLLMKG https://neverblink.eu/ontologies/llm-kg/methods#graphRAG http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#graphRAG http://www.w3.org/2000/01/rdf-schema#label GraphRAG https://neverblink.eu/ontologies/llm-kg/methods#hippoRAG http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#hippoRAG http://www.w3.org/2000/01/rdf-schema#label HippoRAG https://neverblink.eu/ontologies/llm-kg/methods#knowledgeGraphPrompting http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#knowledgeGraphPrompting http://www.w3.org/2000/01/rdf-schema#label Knowledge Graph Prompting (KGP) https://neverblink.eu/ontologies/llm-kg/methods#memWalker http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#memWalker http://www.w3.org/2000/01/rdf-schema#label MemWalker https://neverblink.eu/ontologies/llm-kg/methods#memorizePipeline http://purl.org/dc/terms/subject https://neverblink.eu/ontologies/llm-kg/categories#LLMAugmentedKGConstruction https://neverblink.eu/ontologies/llm-kg/methods#memorizePipeline http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#memorizePipeline http://www.w3.org/2000/01/rdf-schema#comment This pipeline describes the process of constructing and updating the hybrid memory graph from unstructured natural language texts. It leverages LLMs to perform key steps such as formulating vertices and edges, generating hyper-edges, and parsing LLM output into a structured format (e.g., subject-relation-object triples), thereby augmenting KG construction. https://neverblink.eu/ontologies/llm-kg/methods#memorizePipeline http://www.w3.org/2000/01/rdf-schema#label Memorize Pipeline https://neverblink.eu/ontologies/llm-kg/methods#memorizePipeline https://neverblink.eu/ontologies/llm-kg/hasTopCategory https://neverblink.eu/ontologies/llm-kg/top-categories#LLMAugmentedKG https://neverblink.eu/ontologies/llm-kg/methods#mixedRetrievalAlgorithm http://purl.org/dc/terms/subject https://neverblink.eu/ontologies/llm-kg/categories#SynergizedReasoning https://neverblink.eu/ontologies/llm-kg/methods#mixedRetrievalAlgorithm http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#mixedRetrievalAlgorithm http://www.w3.org/2000/01/rdf-schema#comment This algorithm integrates the A*, WaterCircles, and BeamSearch strategies within the QA pipeline to enhance extraction efficacy. By combining the outputs of these distinct traversal methods, it aims to improve overall recall and robustness in retrieving information from the knowledge graph, thus contributing to the synergized reasoning of the system. https://neverblink.eu/ontologies/llm-kg/methods#mixedRetrievalAlgorithm http://www.w3.org/2000/01/rdf-schema#label Mixed Retrieval Algorithm https://neverblink.eu/ontologies/llm-kg/methods#mixedRetrievalAlgorithm https://neverblink.eu/ontologies/llm-kg/hasTopCategory https://neverblink.eu/ontologies/llm-kg/top-categories#SynergizedLLMKG https://neverblink.eu/ontologies/llm-kg/methods#pearl http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#pearl http://www.w3.org/2000/01/rdf-schema#label PEARL https://neverblink.eu/ontologies/llm-kg/methods#personalAiFramework http://purl.org/dc/terms/subject https://neverblink.eu/ontologies/llm-kg/categories#SynergizedKnowledgeRepresentation https://neverblink.eu/ontologies/llm-kg/methods#personalAiFramework http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#personalAiFramework http://www.w3.org/2000/01/rdf-schema#comment This framework introduces a flexible external memory architecture for LLM agents based on a novel hybrid knowledge graph design (supporting standard edges and two types of hyper-edges: thesis and episodic). It enables rich semantic and temporal representations by allowing LLMs to automatically construct and update this structured memory. This aims to synergize LLM and KG capabilities for a unified knowledge representation that enhances personalized, context-aware reasoning. https://neverblink.eu/ontologies/llm-kg/methods#personalAiFramework http://www.w3.org/2000/01/rdf-schema#label PersonalAI Framework https://neverblink.eu/ontologies/llm-kg/methods#personalAiFramework https://neverblink.eu/ontologies/llm-kg/hasTopCategory https://neverblink.eu/ontologies/llm-kg/top-categories#SynergizedLLMKG https://neverblink.eu/ontologies/llm-kg/methods#qaPipeline http://purl.org/dc/terms/subject https://neverblink.eu/ontologies/llm-kg/categories#SynergizedReasoning https://neverblink.eu/ontologies/llm-kg/methods#qaPipeline http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#qaPipeline http://www.w3.org/2000/01/rdf-schema#comment This pipeline handles information search in the memory graph for question answering. It involves an LLM for entity extraction from the natural language question and for generating the final answer, while graph traversal algorithms (like A*, WaterCircles, BeamSearch, and hybrids) are used to retrieve relevant triples from the KG. This exemplifies synergized reasoning by combining LLM capabilities with KG-based structured retrieval. https://neverblink.eu/ontologies/llm-kg/methods#qaPipeline http://www.w3.org/2000/01/rdf-schema#label QA Pipeline https://neverblink.eu/ontologies/llm-kg/methods#qaPipeline https://neverblink.eu/ontologies/llm-kg/hasTopCategory https://neverblink.eu/ontologies/llm-kg/top-categories#SynergizedLLMKG https://neverblink.eu/ontologies/llm-kg/methods#readAgent http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#readAgent http://www.w3.org/2000/01/rdf-schema#label ReadAgent https://neverblink.eu/ontologies/llm-kg/methods#retrievalAugmentedGeneration http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#retrievalAugmentedGeneration http://www.w3.org/2000/01/rdf-schema#label Retrieval-Augmented Generation (RAG) https://neverblink.eu/ontologies/llm-kg/methods#waterCirclesRetrievalAlgorithm http://purl.org/dc/terms/subject https://neverblink.eu/ontologies/llm-kg/categories#SynergizedReasoning https://neverblink.eu/ontologies/llm-kg/methods#waterCirclesRetrievalAlgorithm http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#waterCirclesRetrievalAlgorithm http://www.w3.org/2000/01/rdf-schema#comment This retrieval method is a modified breadth-first search (BFS) algorithm used within the QA pipeline to extract relevant knowledge. It maps query entities to graph vertices, expands outwards, and aggregates triples at intersections, with specific enhancements for handling thesis and episodic hyper-edges, thereby supporting synergized reasoning. https://neverblink.eu/ontologies/llm-kg/methods#waterCirclesRetrievalAlgorithm http://www.w3.org/2000/01/rdf-schema#label WaterCircles Retrieval Algorithm https://neverblink.eu/ontologies/llm-kg/methods#waterCirclesRetrievalAlgorithm https://neverblink.eu/ontologies/llm-kg/hasTopCategory https://neverblink.eu/ontologies/llm-kg/top-categories#SynergizedLLMKG https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/provenance https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/assertion http://www.w3.org/ns/prov#wasAttributedTo https://neverblink.eu/ontologies/llm-kg/agent https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/assertion http://www.w3.org/ns/prov#wasDerivedFrom https://doi.org/10.48550/arXiv.2506.17001 https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/pubinfo https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8 http://purl.org/dc/terms/created 2026-02-26T16:26:56.066Z https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8 http://purl.org/dc/terms/creator https://neverblink.eu/ontologies/llm-kg/agent https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8 http://purl.org/nanopub/x/hasNanopubType https://neverblink.eu/ontologies/llm-kg/PaperAssessmentResult https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8 http://www.w3.org/2000/01/rdf-schema#label LLM-KG assessment for paper 10.48550/arXiv.2506.17001 https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/sig http://purl.org/nanopub/x/hasAlgorithm RSA https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/sig http://purl.org/nanopub/x/hasPublicKey MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAwNz2QK3SEifno78S7+48zUB0xpTex3mAzW73ZimHqNcdEMU5/apslrGrTHGFAt/Chocgo++r6JQp5ygY7NyJHGWdaIqnt85pjX4PbNfLAvapyUO00qZP34fY61w4eZ9UMtleWEsmZKRtQPyJ8ODl46i/rfPuZlcJGpM9Nmy5mpGWuepqIEvF4a/t7pLVeCEDFSYXT+yaiygt6ynIK5f7TtEDhZpeUf/Q74WhMPJXm4yTU/hqOX4IW+50kWHNArGGZwUaXwzyG6M3Zd6UMModryGkLqS4H/MSE3ZA1Ylnms7BfWLEXhMWlaKi6HRV4nGRDLhxVSi9LSRi3LWKLhNIIQIDAQAB https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/sig http://purl.org/nanopub/x/hasSignature Nl1sCLz4NPJ/eReftK+uU8IF4l6RD2AqzioTLZjgXkj250E/pbAnJD3EaWre8KbeH1+nG8OJASvi9/qSLsMR3j7Mz4Mc028xOmFJTOp7vbK0pbR+KlksqAcIaQmHPHc2sAnwAPiVQwsVdPkW8TERN0Fdk5Bq+h02kuuyOZO2fYJHBKI+hhsB4I5g9OCRcOZ8hwlC5G02KnJlqV6Sr0Hu2GauNgcV6+wXPMfZKRqQrNIN7OEm+4k2rONwEXYKT+Pls9uVaOyz/7Nd3bSRD8Nn0kbI+UbSa04ulSDmA98ng0FjAxyLzYDugl7RqzAqm7C4LABYk8aDISnTuTdFIgnorQ== https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/sig http://purl.org/nanopub/x/hasSignatureTarget https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8 https://w3id.org/np/RAyzRXkvyMupy2V4Bj6k0Itf9VTD6Xgx1-X4ijOiyj4a8/sig http://purl.org/nanopub/x/signedBy https://neverblink.eu/ontologies/llm-kg/agent