Empirical study finds overconfidence persists in medical VLMs despite scaling and prompting; post-hoc calibration reduces error while hallucination-aware calibration improves both calibration and AUROC.
A survey of confidence estimation and calibration in large language models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Overconfidence and Calibration in Medical VQA: Empirical Findings and Hallucination-Aware Mitigation
Empirical study finds overconfidence persists in medical VLMs despite scaling and prompting; post-hoc calibration reduces error while hallucination-aware calibration improves both calibration and AUROC.