RLMF uses quality of model self-judgments to refine RL rankings and select training data, achieving SOTA faithful calibration while preserving accuracy and outperforming standard RL by up to 63%.
How to measure metacognition.Frontiers in Human Neuroscience, 8:443, 07 2014
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4verdicts
UNVERDICTED 4representative citing papers
MEDLEY-BENCH reveals an evaluation/control dissociation in AI metacognition where scale improves reflective scoring but not proportional belief revision, with a consistent knowing/doing gap across 35 models.
LLM confidence judgments are dominated by a shared difficulty factor across models, with the confidence-performance link collapsing after removing agreed items, yielding no evidence for individuated metacognition.
Metacognition should serve as a core design principle for AI to improve accuracy, security, and efficiency, demonstrated via a federated learning case study and supported by a new software framework.
citing papers explorer
-
Reinforcement Learning with Metacognitive Feedback Elicits Faithful Uncertainty Expression in LLMs
RLMF uses quality of model self-judgments to refine RL rankings and select training data, achieving SOTA faithful calibration while preserving accuracy and outperforming standard RL by up to 63%.
-
MEDLEY-BENCH: Scale Buys Evaluation but Not Control in AI Metacognition
MEDLEY-BENCH reveals an evaluation/control dissociation in AI metacognition where scale improves reflective scoring but not proportional belief revision, with a consistent knowing/doing gap across 35 models.
-
LLMs Show No Signs Of Individuated Metacognition
LLM confidence judgments are dominated by a shared difficulty factor across models, with the confidence-performance link collapsing after removing agreed items, yielding no evidence for individuated metacognition.
-
Position: Artificial Intelligence Needs Meta Intelligence -- the Case for Metacognitive AI
Metacognition should serve as a core design principle for AI to improve accuracy, security, and efficiency, demonstrated via a federated learning case study and supported by a new software framework.