LLMs show model-specific Trust Elasticity to conflicting evidence on health claims, with near-zero elasticity for false claims and associations to internal uncertainty indicators in open-weight models.
Chen and Roy Ka-Wei Lee , title =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Confident but Conflicted: Internal Uncertainty and Cognitive Dissonance Resolution in LLMs
LLMs show model-specific Trust Elasticity to conflicting evidence on health claims, with near-zero elasticity for false claims and associations to internal uncertainty indicators in open-weight models.