A semantic-graph-enhanced uncertainty model, combining AMR-based entity relations with NLI contradiction scores, improves sentence- and passage-level hallucination detection in LLMs.
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
REJECT 1representative citing papers
citing papers explorer
-
Enhancing Uncertainty Modeling with Semantic Graph for Hallucination Detection
A semantic-graph-enhanced uncertainty model, combining AMR-based entity relations with NLI contradiction scores, improves sentence- and passage-level hallucination detection in LLMs.