Matches in Nanopublications for { ?s ?p ?o <https://w3id.org/np/RALgZ9z8JxiJ1y3hhTAFs3VMu8By-cdlEmFOgcgFoa6zg/assertion>. }
Showing items 1 to 12 of
12
with 100 items per page.
- arXiv.2406.17231 type Entity assertion.
- CogMG type Workflow assertion.
- LiEtAl2023KBBINDER type Workflow assertion.
- CogMG label "CogMG" assertion.
- LiEtAl2023KBBINDER label "KB-BINDER" assertion.
- CogMG comment "CogMG is a collaborative augmentation framework where LLMs act as agents to identify and decompose knowledge requirements that are not covered by the KG. The LLM then completes missing knowledge using its internal parameters, and facilitates the verification and active update of the KG. This process mutually enhances LLM performance in QA by providing factual knowledge and improves KG completeness and alignment with user needs through LLM-driven evolution, demonstrating synergistic reasoning and interaction." assertion.
- LiEtAl2023KBBINDER comment "KB-BINDER (Li et al., 2023) is an existing method for Knowledge Base Question Answering (KBQA) discussed in the related work. It utilizes LLMs' in-context learning to generate draft logical expressions and match executable programs. The paper uses KB-BINDER as an example to contrast CogMG's novelty in addressing scenarios where the knowledge graph lacks necessary information." assertion.
- CogMG subject SynergizedReasoning assertion.
- arXiv.2406.17231 title "CogMG: Collaborative Augmentation Between Large Language Model and Knowledge Graph" assertion.
- arXiv.2406.17231 describes CogMG assertion.
- arXiv.2406.17231 discusses LiEtAl2023KBBINDER assertion.
- CogMG hasTopCategory SynergizedLLMKG assertion.