ChainUQ: Reasoning Consistency-Aware Uncertainty Quantification for Large Language Models

T1Covered by 1 source · first reported by arXiv at 24 Sept, 04:00 UTC

arXiv:2609.26060v1 Announce Type: new Abstract: While large language models (LLMs) exhibit impressive reasoning capabilities, response-level confidence may remain unreliable when intermediate claims conflict with the final conclusion. Therefore, effective uncertainty quantification (UQ) is required…

aiscience

Sources

• T1ChainUQ: Reasoning Consistency-Aware Uncertainty Quantification for Large Language ModelsarXiv24 Sept, 04:00