Reliable but Design-Sensitive: Instrument Uncertainty in LLM Annotation

arXiv:2609.35824 · cs.CL, stat.ME · Submitted 2026-09-22 · Read on arXiv

cs.CL, stat.ME

Submitted: 2026-09-22

Updated: 2026-09-22

Code: https://github.com/soda-lmu/explore_labels

Terminology

• Calibration — 13× in this paper · explained in 7 episodes
• Benchmark — 11× in this paper · explained in 2 episodes
• AI Integration — 4× in this paper · explained in 1 episode
• AUC — 4× in this paper · explained in 2 episodes
• Benchmarking — 3× in this paper · explained in 4 episodes
• Consistency — 2× in this paper · explained in 3 episodes
• Expected Calibration Error — 2× in this paper
• Robustness — 2× in this paper · explained in 5 episodes
• Accuracy — 1× in this paper · explained in 1 episode
• Generative AI — 1× in this paper · explained in 2 episodes
• Large Language Model (LLM) — 1× in this paper · explained in 3 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Representation Learning — 1× in this paper · explained in 2 episodes

Sources

• From Ground Truth to Measurement: A Statistical Framework for Human Labeling
• The threat of analytic flexibility in using large language models to simulate human data
• Prompt Robustness Is Task-Dependent: Comparing Objective and Belief-Style Questions in LLM Evaluation
• BatchGEMBA: Token-Efficient Machine Translation Evaluation with Batched Prompting and Prompt Compression
• What Is Actually Being Annotated? Inter-Prompt Reliability as a Measurement Problem in LLM-Based Social Science Labeling
• Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting
• Scaling Technology Acceptance Analysis with Large Language Model (LLM) Annotation Systems

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering