AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

How to Correctly Report LLM-as-a-Judge Evaluations

Chungpa Lee, Thomas Zeng, Jongwon Jeong, Jy-yong Sohn, Kangwook Lee

ICML 2026 · Large language models › Language, safety and interpretability · approach: Uncertainty quantification for LLMs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
apply Rogan-Gladen bias correction to naive judge scoreLLM-as-a-judge accuracy · Chatbot Arena win rate (GPT-4.1-mini judge)biasImproved
use bias-corrected CI instead of naive p-hat CILLM-as-a-judge accuracy · Monte Carlo (q0,q1)=(0.7,0.9)coverage probability: near-zero coverage -> close to nominal 95%Improved
adaptive calibration allocation (m0 proportional to (1/p-1)sqrt(kappa) m1) vs equal split (ablation)bias-corrected CI · Monte Carlo, m=200CI lengthImproved
bias-correct the estimate before rankingGemini-3-Flash judge · Chatbot Arena rankingexact ranking recovery: 0% -> 37%Improved
bias-correct the estimate before rankingClaude-Haiku-4.5 judge · Chatbot Arena rankingexact ranking recovery: 7% -> 40%Improved
bias-correct the estimate before ranking (10% calibration)GPT-4.1 judge · Chatbot Arena rankingKendall's tau: 0.981 -> 0.915Hurt
increase calibration fraction from 10% to 40% (ablation)bias-corrected estimator, GPT-4.1 judge · Chatbot Arena rankingexact ranking recovery: 46% -> 75%Improved
use difference/calibration-only/conditional calibration estimators under label distribution shiftexisting bias-correction estimators · Monte Carlo distribution shiftbiasHurt
use misclassification-adjusted estimator under label distribution shiftbias-corrected estimator · Monte Carlo distribution shiftbiasImproved
use calibrated LLM judge instead of human-only evaluation (current judges on Chatbot Arena)bias-corrected estimator · Chatbot ArenavarianceHurt
use GPT-4.1-mini judge on AlpacaEval instead of Chatbot Arenabias-corrected estimator · AlpacaEvalvariance vs human-onlyImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy