The field of knowledge graphs and large language models is rapidly evolving, with a focus on improving the representation and reasoning capabilities of these models. Recent developments have highlighted the importance of integrating symbolic and contextual knowledge to enable more effective semantic transfer and reasoning. The use of residual quantization and masked diffusion models has shown promise in bridging the gap between knowledge graph embeddings and large language models, allowing for more seamless fusion of structured and unstructured knowledge. Additionally, the application of large language models to tasks such as drug repurposing, biomedical concept representation, and knowledge graph construction has demonstrated significant potential for advancing research in these areas. Noteworthy papers include ReaLM, which proposes a novel framework for bridging the gap between knowledge graph embeddings and large language models, and Knowledge Reasoning Language Model, which achieves unified coordination between large language model knowledge and knowledge graph context for inductive knowledge graph reasoning.
Advances in Knowledge Graphs and Large Language Models
Sources
ReaLM: Residual Quantization Bridging Knowledge Graph Embeddings and Large Language Models
Scalable and Explainable Enterprise Knowledge Discovery Using Graph-Centric Hybrid Retrieval
A Large-Language-Model Assisted Automated Scale Bar Detection and Extraction Framework for Scanning Electron Microscopic Images
Are Large Language Models Effective Knowledge Graph Constructors?
Medical Interpretability and Knowledge Maps of Large Language Models
Unifying Deductive and Abductive Reasoning in Knowledge Graphs with Masked Diffusion Model
Ontolearn-A Framework for Large-scale OWL Class Expression Learning in Python
VizCopilot: Fostering Appropriate Reliance on Enterprise Chatbots with Context Visualization
KnowledgeTrail: Generative Timeline for Exploration and Sensemaking of Historical Events and Knowledge Formation
From Knowledge to Treatment: Large Language Model Assisted Biomedical Concept Representation for Drug Repurposing
Semantic knowledge guides innovation and drives cultural evolution
Smart UX-design for Rescue Operations Wearable - A Knowledge Graph Informed Visualization Approach for Information Retrieval in Emergency Situations
The Past Still Matters: A Temporally-Valid Data Discovery System
Meronymic Ontology Extraction via Large Language Models
Interpreting the Latent Structure of Operator Precedence in Language Models
Knowledge Reasoning Language Model: Unifying Knowledge and Language for Inductive Knowledge Graph Reasoning
VisAider: AI-Assisted Context-Aware Visualization Support for Data Presentations