LLM Judges as Raters: A Pre-Registered Audit of Severity, Halo, Reliability, and Version Instability in LLM Essay Scoring on Public Corpora

arXiv:2608.29517 · cs.CL, cs.CY · Submitted 2026-08-30 · Read on arXiv

cs.CL, cs.CY

Submitted: 2026-08-30

Updated: 2026-08-30

Terminology

• Calibration — 24× in this paper · explained in 7 episodes
• LLM judge — 14× in this paper · explained in 2 episodes
• Robustness — 8× in this paper · explained in 5 episodes
• Consistency — 6× in this paper · explained in 3 episodes
• Generalizability — 6× in this paper · explained in 6 episodes
• Accuracy — 5× in this paper · explained in 1 episode
• Benchmark — 5× in this paper · explained in 2 episodes
• Self-consistency — 5× in this paper · explained in 2 episodes
• LLM-as-a-Judge — 3× in this paper · explained in 3 episodes
• Distribution Shift — 2× in this paper · explained in 6 episodes
• Large Language Models (LLMs) — 2× in this paper · explained in 6 episodes
• Automated Student Assessment Prize — 1× in this paper
• Quantization — 1× in this paper · explained in 12 episodes
• Residualization — 1× in this paper · explained in 2 episodes

Sources

• Towards Self-Referential Analytic Assessment: A Profile-Based Approach to L2 Writing Evaluation with LLMs
• Correcting Human Labels for Rater Effects in AI Evaluation: An Item Response Theory Approach
• Estimating LLM Grading Ability and Response Difficulty in Automatic Short Answer Grading via Item Response Theory
• LLMs Do Not Grade Essays Like Humans
• Agreement Metrics for LLM-as-Judge Evaluation: What to Report and Why
• Exploring LLM Autoscoring Reliability in Large-Scale Writing Assessments Using Generalizability Theory
• LLM Judges Have Dark Current: A Psychometric Datasheet for LLM-as-a-Judge Evaluation
• Where Does the Noise Come From? A Variance-Components Decomposition of Non-Determinism in LLM Brand Answers

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