Knowledge Tree
Relations (1)
related 3.17 — strongly supporting 8 facts
Large Language Models are increasingly utilized to perform tasks on temporal knowledge graphs, such as forecasting {fact:2, fact:5, fact:6} and question answering {fact:4, fact:8}. Research explores zero-shot relational learning [1] and in-context learning [2] to bridge these domains, despite existing challenges regarding scalability and modeling complexity [3].
Facts (8)
Sources
Practices, opportunities and challenges in the fusion of knowledge ... frontiersin.org 5 facts
claimTemporal knowledge graphs are rarely combined with large language models due to scalability concerns and complex modeling requirements, as noted by Wang et al. (2023b).
referenceThe paper 'Two-stage generative question answering on temporal knowledge graph using large language models' (arXiv:2402.16568) proposes a two-stage generative approach for question answering over temporal knowledge graphs using large language models.
referenceR. Liao, X. Jia, Y. Li, Y. Ma, and V. Tresp published 'Gentkg: Generative forecasting on temporal knowledge graph with large language models' as an arXiv preprint in 2023.
referenceR. Liao, X. Jia, Y. Li, Y. Ma, and V. Tresp published 'Gentkg: generative forecasting on temporal knowledge graph with large language models' in the Findings of the Association for Computational Linguistics: NAACL 2024.
referenceThe paper 'ZRLLM: zero-shot relational learning on temporal knowledge graphs with large language models' (Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics) presents a zero-shot approach for relational learning on temporal knowledge graphs using large language models.
Combining Knowledge Graphs and Large Language Models - arXiv arxiv.org 2 facts
referenceLee et al. demonstrated that LLMs can learn patterns from historical data in Temporal Knowledge Graphs using in-context learning (ICL) without requiring special architectures or modules.
claimLLMs can perform forecasting using Temporal Knowledge Graphs (TKGs), which are a subset of Knowledge Graphs containing directions and timestamps.
Large Language Models Meet Knowledge Graphs for Question ... arxiv.org 1 fact
referenceGao et al. (2024) developed a two-stage generative question answering method on temporal knowledge graphs using large language models, published in the ACL Findings proceedings.