The field of natural language processing is moving towards tighter integration of knowledge graphs and large language models. Recent research has focused on developing innovative methods for incorporating knowledge graphs into large language models to improve their performance and accuracy. One of the key directions is the development of lightweight and efficient frameworks that can leverage the full potential of large language models to tackle complex reasoning tasks. Another important area of research is the development of scalable predictive modeling approaches to identify duplicate adverse event reports for drugs and vaccines, which can help improve pharmacovigilance. The use of crowdsourcing-based knowledge graph construction and post-training language models for continual relation extraction are also noteworthy trends. Notable papers include LightPROF, which proposes a novel lightweight reasoning framework for large language models on knowledge graphs, and A Scalable Predictive Modelling Approach to Identifying Duplicate Adverse Event Reports for Drugs and Vaccines, which presents a new model that achieves higher precision and recall for duplicate detection. Additionally, the survey paper on circuit foundation models provides a comprehensive overview of the latest progress in this emerging research trend.
Advances in Integrating Knowledge Graphs and Large Language Models
Sources
LightPROF: A Lightweight Reasoning Framework for Large Language Model on Knowledge Graph
A Scalable Predictive Modelling Approach to Identifying Duplicate Adverse Event Reports for Drugs and Vaccines
A Survey of Circuit Foundation Model: Foundation AI Models for VLSI Circuit Design and EDA
Practical Poisoning Attacks against Retrieval-Augmented Generation
QE-RAG: A Robust Retrieval-Augmented Generation Benchmark for Query Entry Errors
Crowdsourcing-Based Knowledge Graph Construction for Drug Side Effects Using Large Language Models with an Application on Semaglutide
Post-Training Language Models for Continual Relation Extraction
BRIDGES: Bridging Graph Modality and Large Language Models within EDA Tasks
Evaluating Knowledge Graph Based Retrieval Augmented Generation Methods under Knowledge Incompleteness
GraphRAFT: Retrieval Augmented Fine-Tuning for Knowledge Graphs on Graph Databases
Knowledge-Instruct: Effective Continual Pre-training from Limited Data using Instructions
From Superficial to Deep: Integrating External Knowledge for Follow-up Question Generation Using Knowledge Graph and LLM
Open Problems and a Hypothetical Path Forward in LLM Knowledge Paradigms
KG-LLM-Bench: A Scalable Benchmark for Evaluating LLM Reasoning on Textualized Knowledge Graphs