GENUINE uses dependency parse trees and learnable graph pooling to produce uncertainty scores for LLM outputs, claiming AUROC gains of up to 29% over semantic entropy baselines.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
GENUINE: Graph Enhanced Multi-level Uncertainty Estimation for Large Language Models
GENUINE uses dependency parse trees and learnable graph pooling to produce uncertainty scores for LLM outputs, claiming AUROC gains of up to 29% over semantic entropy baselines.