Image de couverture de livre

Screen Before You Interpret: A Portable Validity Protocol for Benchmark-Based LLM Confidence Signals

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteur principal:	Cacioli, Jon-Paul
Format:	Preprint
Publié:	2026
Sujets:	Computation and Language Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Before You Interpret the Profile: Validity Scaling for LLM Metacognitive Self-Report
par: Cacioli, Jon-Paul
Publié: (2026)

Concurrent Criterion Validation of a Validity Screen for LLM Confidence Signals via Selective Prediction
par: Cacioli, Jon-Paul
Publié: (2026)

Verbal Confidence Saturation in 3-9B Open-Weight Instruction-Tuned LLMs: A Pre-Registered Psychometric Validity Screen
par: Cacioli, Jon-Paul
Publié: (2026)

LLMs as Signal Detectors: Sensitivity, Bias, and the Temperature-Criterion Analogy
par: Cacioli, Jon-Paul
Publié: (2026)

Instruction Complexity Induces Positional Collapse in Adversarial LLM Evaluation
par: Cacioli, Jon-Paul
Publié: (2026)

Beyond the Mean: Within-Model Reliable Change Detection for LLM Evaluation
par: Cacioli, Jon-Paul
Publié: (2026)

Distilling Self-Consistency into Verbal Confidence: A Pre-Registered Negative Result and Post-Hoc Rescue on Gemma 3 4B
par: Cacioli, Jon-Paul
Publié: (2026)

Do LLMs Know What They Know? Measuring Metacognitive Efficiency with Signal Detection Theory
par: Cacioli, Jon-Paul
Publié: (2026)

Domain-level metacognitive monitoring in frontier LLMs: A 33-model atlas
par: Cacioli, Jon-Paul
Publié: (2026)

Weber's Law in Transformer Magnitude Representations: Efficient Coding, Representational Geometry, and Psychophysical Laws in Language Models
par: Cacioli, Jon-Paul
Publié: (2026)

Below-Chance Blindness: Prompted Underperformance in Small LLMs Produces Positional Bias Rather than Answer Avoidance
par: Cacioli, Jon-Paul
Publié: (2026)

Option-Order Randomisation Reveals a Distributional Position Attractor in Prompted Sandbagging
par: Cacioli, Jon-Paul
Publié: (2026)

Cross-Entropy Is Load-Bearing: A Pre-Registered Scope Test of the K-Way Energy Probe on Bidirectional Predictive Coding
par: Cacioli, Jon-Paul
Publié: (2026)

Exemplar Retrieval Without Overhypothesis Induction: Limits of Distributional Sequence Learning in Early Word Learning
par: Cacioli, Jon-Paul
Publié: (2026)

Categorical Perception in Large Language Model Hidden States: Structural Warping at Digit-Count Boundaries
par: Cacioli, Jon-Paul
Publié: (2026)

The Metacognitive Monitoring Battery: A Cross-Domain Benchmark for LLM Self-Monitoring
par: Cacioli, Jon-Paul
Publié: (2026)

Look Before You Leap: Autonomous Exploration for LLM Agents
par: Ye, Ziang, et autres
Publié: (2026)

SnapKV: LLM Knows What You are Looking for Before Generation
par: Li, Yuhong, et autres
Publié: (2024)

Think Twice Before You Write -- an Entropy-based Decoding Strategy to Enhance LLM Reasoning
par: He, Jiashu, et autres
Publié: (2026)

Read Before You Think: Mitigating LLM Comprehension Failures with Step-by-Step Reading
par: Han, Feijiang, et autres
Publié: (2025)

Do Before You Judge: Self-Reference as a Pathway to Better LLM Evaluation
par: Lin, Wei-Hsiang, et autres
Publié: (2025)

Verify Before You Commit: Towards Faithful Reasoning in LLM Agents via Self-Auditing
par: Yuan, Wenhao, et autres
Publié: (2026)

Look Before You Leap: Enhancing Attention and Vigilance Regarding Harmful Content with GuidelineLLM
par: Zhang, Shaoqing, et autres
Publié: (2024)

Cite Before You Speak: Enhancing Context-Response Grounding in E-commerce Conversational LLM-Agents
par: Zeng, Jingying, et autres
Publié: (2025)

Confidence Estimation for LLM-Based Dialogue State Tracking
par: Sun, Yi-Jyun, et autres
Publié: (2024)

Quantisation Reshapes the Metacognitive Geometry of Language Models
par: Cacioli, Jon-Paul
Publié: (2026)

ProtocolLLM: RTL Benchmark for SystemVerilog Generation of Communication Protocols
par: Sheth, Arnav, et autres
Publié: (2025)

Silicon Bureaucracy and AI Test-Oriented Education: Contamination Sensitivity and Score Confidence in LLM Benchmarks
par: Song, Yiliang, et autres
Publié: (2026)

Confident or Seek Stronger: Exploring Uncertainty-Based On-device LLM Routing From Benchmarking to Generalization
par: Chuang, Yu-Neng, et autres
Publié: (2025)

Same Geometry, Opposite Noise: Transformer Magnitude Representations Lack Scalar Variability
par: Cacioli, Jon-Paul
Publié: (2026)

Think Thrice Before You Act: Progressive Thought Refinement in Large Language Models
par: Du, Chengyu, et autres
Publié: (2024)

Warm Up Before You Train: Unlocking General Reasoning in Resource-Constrained Settings
par: Shrestha, Safal, et autres
Publié: (2025)

Contrastive Attribution in the Wild: An Interpretability Analysis of LLM Failures on Realistic Benchmarks
par: Tan, Rongyuan, et autres
Publié: (2026)

From Confidence to Collapse in LLM Factual Robustness
par: Fastowski, Alina, et autres
Publié: (2025)

Think Before You Lie: How Reasoning Leads to Honesty
par: Yuan, Ann, et autres
Publié: (2026)

Think Before You Act: Decision Transformers with Working Memory
par: Kang, Jikun, et autres
Publié: (2023)

Confidence and Calibration of Activation Oracles for Reliable Interpretation of Language Model Internals
par: Torrielli, Federico, et autres
Publié: (2026)

Look Before You Leap: Problem Elaboration Prompting Improves Mathematical Reasoning in Large Language Models
par: Liao, Haoran, et autres
Publié: (2024)

Stop Before You Fail: Operational Capability Boundaries for Mitigating Unproductive Reasoning in Large Reasoning Models
par: Zhang, Qingjie, et autres
Publié: (2025)

Think Before You Speak: Cultivating Communication Skills of Large Language Models via Inner Monologue
par: Zhou, Junkai, et autres
Publié: (2023)