https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/Head https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs http://www.nanopub.org/nschema#hasAssertion https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/assertion https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs http://www.nanopub.org/nschema#hasProvenance https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/provenance https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs http://www.nanopub.org/nschema#hasPublicationInfo https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/pubinfo https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://www.nanopub.org/nschema#Nanopublication https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/assertion https://doi.org/10.48550/arXiv.2510.17934 http://purl.org/dc/terms/title ATLASKV: AUGMENTING LLMS WITH BILLION-SCALE KNOWLEDGE GRAPHS IN 20GB VRAM https://doi.org/10.48550/arXiv.2510.17934 http://purl.org/spar/cito/describes https://neverblink.eu/ontologies/llm-kg/methods#AtlasKV https://doi.org/10.48550/arXiv.2510.17934 http://purl.org/spar/cito/describes https://neverblink.eu/ontologies/llm-kg/methods#HiKVP https://doi.org/10.48550/arXiv.2510.17934 http://purl.org/spar/cito/describes https://neverblink.eu/ontologies/llm-kg/methods#KG2KV https://doi.org/10.48550/arXiv.2510.17934 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#InContextLearning https://doi.org/10.48550/arXiv.2510.17934 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#KBLaM https://doi.org/10.48550/arXiv.2510.17934 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#RetrievalAugmentedGeneration https://doi.org/10.48550/arXiv.2510.17934 http://purl.org/spar/cito/discusses https://neverblink.eu/ontologies/llm-kg/methods#ZeroShotLearning https://doi.org/10.48550/arXiv.2510.17934 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#AtlasKV http://purl.org/dc/terms/subject https://neverblink.eu/ontologies/llm-kg/categories#KGEnhancedLLMInference https://neverblink.eu/ontologies/llm-kg/methods#AtlasKV http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#AtlasKV http://www.w3.org/2000/01/rdf-schema#comment AtlasKV is a parametric knowledge integration method that augments LLMs with billion-scale KGs. It is designed to overcome limitations of existing methods by improving LLM performance in terms of knowledge grounding, generalization, and scalability during the inference stage, without requiring external retrievers or retraining. https://neverblink.eu/ontologies/llm-kg/methods#AtlasKV http://www.w3.org/2000/01/rdf-schema#label AtlasKV https://neverblink.eu/ontologies/llm-kg/methods#AtlasKV https://neverblink.eu/ontologies/llm-kg/hasTopCategory https://neverblink.eu/ontologies/llm-kg/top-categories#KGEnhancedLLM https://neverblink.eu/ontologies/llm-kg/methods#HiKVP http://purl.org/dc/terms/subject https://neverblink.eu/ontologies/llm-kg/categories#KGEnhancedLLMInference https://neverblink.eu/ontologies/llm-kg/methods#HiKVP http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#HiKVP http://www.w3.org/2000/01/rdf-schema#comment HiKVP (Hierarchical Key-Value Pruning) is an algorithm that dramatically reduces computational and memory overhead during LLM inference by hierarchically clustering and pruning KGKVs. It maintains high knowledge grounding accuracy while enabling scalable integration of billion-scale KGs into LLMs at inference time. https://neverblink.eu/ontologies/llm-kg/methods#HiKVP http://www.w3.org/2000/01/rdf-schema#label HiKVP https://neverblink.eu/ontologies/llm-kg/methods#HiKVP https://neverblink.eu/ontologies/llm-kg/hasTopCategory https://neverblink.eu/ontologies/llm-kg/top-categories#KGEnhancedLLM https://neverblink.eu/ontologies/llm-kg/methods#InContextLearning http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#InContextLearning http://www.w3.org/2000/01/rdf-schema#label In-context learning https://neverblink.eu/ontologies/llm-kg/methods#KBLaM http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#KBLaM http://www.w3.org/2000/01/rdf-schema#label KBLaM https://neverblink.eu/ontologies/llm-kg/methods#KG2KV http://purl.org/dc/terms/subject https://neverblink.eu/ontologies/llm-kg/categories#KGEnhancedLLMPretraining https://neverblink.eu/ontologies/llm-kg/methods#KG2KV http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#KG2KV http://www.w3.org/2000/01/rdf-schema#comment KG2KV is a pipeline that transforms KG triples into high-quality Q-K-V data, which serves as training data for LLMs. This method enhances the generalization performance and efficient knowledge integration by enabling better injection of KGs into LLMs' parametric representations, thus improving their knowledge expression. https://neverblink.eu/ontologies/llm-kg/methods#KG2KV http://www.w3.org/2000/01/rdf-schema#label KG2KV https://neverblink.eu/ontologies/llm-kg/methods#KG2KV https://neverblink.eu/ontologies/llm-kg/hasTopCategory https://neverblink.eu/ontologies/llm-kg/top-categories#KGEnhancedLLM 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 https://neverblink.eu/ontologies/llm-kg/methods#ZeroShotLearning http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://purl.org/spar/fabio/Workflow https://neverblink.eu/ontologies/llm-kg/methods#ZeroShotLearning http://www.w3.org/2000/01/rdf-schema#label Zero-shot learning https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/provenance https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/assertion http://www.w3.org/ns/prov#wasAttributedTo https://neverblink.eu/ontologies/llm-kg/agent https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/assertion http://www.w3.org/ns/prov#wasDerivedFrom https://doi.org/10.48550/arXiv.2510.17934 https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/pubinfo https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs http://purl.org/dc/terms/created 2026-02-26T16:25:34.999Z https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs http://purl.org/dc/terms/creator https://neverblink.eu/ontologies/llm-kg/agent https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs http://purl.org/nanopub/x/hasNanopubType https://neverblink.eu/ontologies/llm-kg/PaperAssessmentResult https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs http://www.w3.org/2000/01/rdf-schema#label LLM-KG assessment for paper 10.48550/arXiv.2510.17934 https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/sig http://purl.org/nanopub/x/hasAlgorithm RSA https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/sig http://purl.org/nanopub/x/hasPublicKey MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAwNz2QK3SEifno78S7+48zUB0xpTex3mAzW73ZimHqNcdEMU5/apslrGrTHGFAt/Chocgo++r6JQp5ygY7NyJHGWdaIqnt85pjX4PbNfLAvapyUO00qZP34fY61w4eZ9UMtleWEsmZKRtQPyJ8ODl46i/rfPuZlcJGpM9Nmy5mpGWuepqIEvF4a/t7pLVeCEDFSYXT+yaiygt6ynIK5f7TtEDhZpeUf/Q74WhMPJXm4yTU/hqOX4IW+50kWHNArGGZwUaXwzyG6M3Zd6UMModryGkLqS4H/MSE3ZA1Ylnms7BfWLEXhMWlaKi6HRV4nGRDLhxVSi9LSRi3LWKLhNIIQIDAQAB https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/sig http://purl.org/nanopub/x/hasSignature kI8bqlYDgCuCaY9kYe5Ri7LxFScfd38o3qzlVJ2Hrl3u3GxHKcVsTgwktXxMpDAxCPmymjiBiu5Oz788n+bV+y9PzFaK5HZHgSX/xma78Fm3cVElSwVxKQCuGgv/o96v0WbP4BmlJIukJN/pKvV3Wodd0YKwXty2zKymN9PKjIUNLZ7zsePmEGi9Cbi8JWTV1QJodzHXyaDvuwsIOY19p6PiiCSlY1sQVIFk0X7FZDGXO30oD6IzxCiD0Jdgj1s/pPrc7vDVxPc5Y6VSN4VG5S/I9+nAJ6MEqtzndBzBoQ+MN0r43MAPqgnqYJy3Et2Q4KnjHquCRj8ffK0/l45GHw== https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/sig http://purl.org/nanopub/x/hasSignatureTarget https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs https://w3id.org/np/RAaGcIMVdjU39we9S6HyZATb-2KUg6GD1QQrFmMRlcSxs/sig http://purl.org/nanopub/x/signedBy https://neverblink.eu/ontologies/llm-kg/agent