Jon-Paul Cacioli

1. Domain-level metacognitive monitoring in frontier LLMs: A 33-model atlas
   2026/04/21 by Jon-Paul Cacioli · 1 voice
   Computer Science · #cs.AI #cs.CL #cs.LG
2. LLMs as Signal Detectors: Sensitivity, Bias, and the Temperature-Criterion Analogy
   2026/03/16 by Jon-Paul Cacioli · 1 voice · 2 citations
   Computer Science · #cs.CL #cs.AI
3. Do LLMs Know What They Know? Measuring Metacognitive Efficiency with Signal Detection Theory
   2026/03/26 by Jon-Paul Cacioli · 1 voice · 1 citation
   Computer Science · #cs.AI #cs.CL
4. Categorical Perception in Large Language Model Hidden States: Structural Warping at Digit-Count Boundaries
   2026/03/30 by Jon-Paul Cacioli · 1 voice
   Computer Science · #cs.AI #cs.CL
5. Exemplar Retrieval Without Overhypothesis Induction: Limits of Distributional Sequence Learning in Early Word Learning
   2026/04/06 by Jon-Paul Cacioli · 1 voice
   Computer Science · Neuroscience · Psychology · #Autoregressive model #Child and Animal Learning Development #Equivalence (formal languages) #Feature (linguistics) #Feature learning #Language Development and Disorders #Neurobiology of Language and Bilingualism #Sequence (biology) #Sequence labeling #Transformer #Word (group theory) #cs.AI #cs.CL
6. Same Geometry, Opposite Noise: Transformer Magnitude Representations Lack Scalar Variability
   2026/04/06 by Jon-Paul Cacioli · 1 voice
   Biochemistry, Genetics and Molecular Biology · Computer Science · #cs.CL #q-bio.QM
7. Distilling Self-Consistency into Verbal Confidence: A Pre-Registered Negative Result and Post-Hoc Rescue on Gemma 3 4B
   2026/04/27 by Jon-Paul Cacioli · 1 voice
   Computer Science · #cs.AI #cs.CL