Jon-Paul Cacioli
Identifiers
No identifiers captured yet.
Papers (17)
- Beyond the Mean: Within-Model Reliable Change Detection for LLM Evaluation cs.CL · 2026 · author #1
- Instruction Complexity Induces Positional Collapse in Adversarial LLM Evaluation cs.CL · 2026 · author #1
- Option-Order Randomisation Reveals a Distributional Position Attractor in Prompted Sandbagging cs.CL · 2026 · author #1
- Below-Chance Blindness: Prompted Underperformance in Small LLMs Produces Positional Bias Rather than Answer Avoidance cs.CL · 2026 · author #1
- Distilling Self-Consistency into Verbal Confidence: A Pre-Registered Negative Result and Post-Hoc Rescue on Gemma 3 4B cs.CL · 2026 · author #1
- Verbal Confidence Saturation in 3-9B Open-Weight Instruction-Tuned LLMs: A Pre-Registered Psychometric Validity Screen cs.CL · 2026 · author #1
- Cross-Entropy Is Load-Bearing: A Pre-Registered Scope Test of the K-Way Energy Probe on Bidirectional Predictive Coding cs.CL · 2026 · author #1
- Domain-level metacognitive monitoring in frontier LLMs: A 33-model atlas cs.CL · 2026 · author #1
- Concurrent Criterion Validation of a Validity Screen for LLM Confidence Signals via Selective Prediction cs.CL · 2026 · author #1
- Screen Before You Interpret: A Portable Validity Protocol for Benchmark-Based LLM Confidence Signals cs.CL · 2026 · author #1
- Before You Interpret the Profile: Validity Scaling for LLM Metacognitive Self-Report cs.CL · 2026 · author #1
- The Metacognitive Monitoring Battery: A Cross-Domain Benchmark for LLM Self-Monitoring cs.CL · 2026 · author #1
- K-Way Energy Probes for Metacognition Reduce to Softmax in Discriminative Predictive Coding Networks cs.LG · 2026 · author #1
- Quantisation Reshapes the Metacognitive Geometry of Language Models cs.CL · 2026 · author #1
- Exemplar Retrieval Without Overhypothesis Induction: Limits of Distributional Sequence Learning in Early Word Learning cs.CL · 2026 · author #1
- Same Geometry, Opposite Noise: Transformer Magnitude Representations Lack Scalar Variability cs.CL · 2026 · author #1
- Categorical Perception in Large Language Model Hidden States: Structural Warping at Digit-Count Boundaries cs.CL · 2026 · author #1
Mentions
No mention provenance yet.