-
Review highlights key biases in observational studies
A critical examination of validity threats in observational studies, including confounding, selection bias, time-varying confounding, measurement error, and missing data handling strategies.
-
CORE: Data Augmentation for Link Prediction via Information Bottleneck
CORE applies Information Bottleneck principles to augment graph data for link prediction, simultaneously recovering missing edges and reducing noise to enhance model robustness.
-
Prompt-driven KG-enhanced LLM reasoning improved KBQA accuracy
Prompt-driven framework combining LLMs with knowledge graphs for reliable knowledge-based question answering through structured subgraph retrieval and stepwise reasoning validation.
