Framework that uses LLMs to generate study triplets and triplet loss to learn embeddings for clustering observational studies prior to meta-analysis, applied to 58 studies on preterm birth cognitive outcomes.
hub
A re -evaluation of random-effects meta-analysis
2 Pith papers cite this work, alongside 2,509 external citations. Polarity classification is still indexing.
2
Pith papers citing it
2,509
external citations · external index
hub tools
fields
stat.ME 2verdicts
UNVERDICTED 2representative citing papers
Proposes and compares CG-C, 2MA-C and MIX-C random-effects approaches for clustered flexible calibration plots, with simulation and case-study evidence favoring 2MA-C (splines) for overall curves and MIX-C for cluster-specific curves.
citing papers explorer
-
Learning study similarity to investigate heterogeneity in meta-analysis using LLMs and triplet loss
Framework that uses LLMs to generate study triplets and triplet loss to learn embeddings for clustering observational studies prior to meta-analysis, applied to 58 studies on preterm birth cognitive outcomes.
-
Clustered Flexible Calibration Plots For Binary Outcomes Using Random Effects Modeling
Proposes and compares CG-C, 2MA-C and MIX-C random-effects approaches for clustered flexible calibration plots, with simulation and case-study evidence favoring 2MA-C (splines) for overall curves and MIX-C for cluster-specific curves.