Bias in LLM-as-a-Judge

Follow this topic Ask about this topic

This field examines how automated judges can favor certain answers for reasons unrelated to quality, including response order, writing style, verbosity, model identity, language, and demographic or cultural cues. Researchers use controlled counterfactual tests and benchmarks such as MT-Bench, RewardBench, and multilingual or domain-specific evaluation sets to measure when judgments change and whether they match human ratings. Mitigation approaches include randomizing presentation, adding rubrics and reference answers, calibrating scores with human comparisons, combining judges robustly, and training models to separate relevant evidence from distracting cues. A central challenge is that repeatable judgments and agreement among models do not guarantee validity, while interventions that reduce bias can also erase real quality differences or legitimate differences in human preferences.

Latest papers

118 papers21 papers in the last four weeks, up 11% on the four weeks before. 0.1% of all new papers.

Ask the Right Comparison:Bias-Aware Bayesian Active Top- $k$ Ranking with LLM Judges
Jul 2, 2026Jian Xu, Delu Zeng, John Paisley +1Pairwise Comparison LLM-as-a-Judge
RIKEN iTHEMS · RIKEN AIP · South China University of Technology +1
RoPoLL: Robust Panel of LLM Judges
Jun 29, 2026Anish Acharya, Kris W Pan, Brian VerkhovskyLLM Evaluation LLM-as-a-Judge
Amazon Web Services
BabelJudge: Measuring LLM-as-a-Judge Reliability Across Languages and Agent Trajectories
Jun 21, 2026Shreyas KCMultilingual Language Model Evaluation LLM-as-a-Judge
Reliability without Validity: A Systematic, Large-Scale Evaluation of LLM-as-a-Judge Models Across Agreement, Consistency, and Bias
Jun 17, 2026Justin D. Norman, Michael U. Rivera, D. Alex HughesInter-Rater Reliability LLM Evaluation
UC Berkeley School of Information 102 South Hall Drive, Berkeley CA, 94708
Quantifying and Auditing LLM Evaluation via Positive--Unlabeled Learning
Jun 17, 2026Zilong Zhang, Yi-Ting Hung, Lei Ding +1Inverse Optimal Transport LLM-as-a-Judge
Department of Mathematics and Statistics Georgia State University · Department of Statistics University of Manitoba
Evaluating Second-Order Bias of LLMs Through Epistemic Entitlement
Jun 16, 2026Ramaravind Kommiya Mothilal, Terry Jingchen Zhang, Raiyan Ahmed +3LLM-as-a-Judge Social Bias in Language Models
University of Toronto · Vector Institute · EuroSafeAI +1
LLM Judges Have Dark Current: Stratified Null and Utility Tests for LLM Judge Reliability
Jun 14, 2026Hiroyasu Usami, Keisuke Hara, Ayato Tsuboi +1LLM Evaluation LLM-as-a-Judge
Department of Computer Science, Graduate School of Engineering, Chubu University, Kasugai, Aichi 487-8501, Japan · Mitsubishi Heavy Industries, Ltd., Research & Innovation Center, Heat Transfer Research Department, Takasago, Hyogo 676-8686, Japan
Jury Duty: Calibration and Orientation Failures in MLLM-as-a-Judge Under Cultural Ambiguity
Jun 12, 2026Daniel Lee, Harsh Sharma, Eunkyu Park +5LLM-as-a-Judge Language Model Calibration
Salesforce AI Research · University of Colorado Boulder · Carnegie Mellon University +2
Does the Judge Prefer English? Evaluating Language-Switching Invariance in LLM-as-a-Judge
Jun 12, 2026Shaojie YinMultilingual Language Model Evaluation LLM Evaluation
Shanghai Artificial Intelligence Laboratory Shanghai, China
Self- and Other-Labels Induce Bidirectional Bias in LLM Judges
Jun 6, 2026Songeun Chae, Min Kim, Donghoon Jung +2Bias in LLM-as-a-Judge Language Model Bias Evaluation
School of Digital Humanities and Computational Social Sciences, KAIST, South Korea
Turning Bias into Bugs: Bandit-Guided Style Manipulation Attacks on LLM Judges
May 24, 2026Xianglin Yang, Bryan Hooi, Gelei Deng +2LLM Evaluation LLM-as-a-Judge
School of Computing, National University of Singapore, Singapore · Nanyang Technological University, Singapore
AMEL: Accumulated Message Effects on LLM Judgments
May 21, 2026Sid-Ali TemkitLLM Evaluation Language Model Generation Evaluation
Faithful or Fabricated? A Causal Framework for Rationalization Bias in LLM Judges
May 13, 2026Riya Tapwal, Abhishek Kumar, Carsten MapleLLM-as-a-Judge Faithfulness of Language Model Explanations
Indian Institute of Technology (IIT) Mandi · The Alan Turing Institute, London, U.K. · Warwick Manufacturing Group University of Warwick, U.K.
Two Ways to De-Bias an LLM-as-a-Judge: A Continuous-Score Comparison of Hierarchical Bayesian Calibration and Neural-ODE Score Transport
May 9, 2026Andrea MorandiPost-Hoc Calibration LLM-as-a-Judge
Cisco Systems, Inc.
Bias and Uncertainty in LLM-as-a-Judge Estimation
May 7, 2026James FiedlerLLM Evaluation LLM-as-a-Judge
Indeed Inc.
Prosa: Rubric-Based Evaluation of LLMs on Real User Chats in Brazilian Portuguese
May 2, 2026Roseval Malaquias Junior, Giovana Kerche Bonás, Thales Sales Almeida +6LLM Evaluation LLM-as-a-Judge
Maritaca AI · Jusbrasil
Judging the Judges: A Systematic Evaluation of Bias Mitigation Strategies in LLM-as-a-Judge Pipelines
Apr 25, 2026Sadman Kabir SoumikDebiased ML LLM-as-a-Judge
Independent Researcher
Quantifying and Mitigating Self-Preference Bias of LLM Judges
Apr 24, 2026Jinming Yang, Zheng Hu, Chuxian Qiu +3LLM-as-a-Judge Language Model Bias Evaluation
CompleX Lab, School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, China
The Coin Flip Judge? Reliability and Bias in LLM-as-a-Judge Evaluation
Apr 23, 2026Abel YagubyanInter-Rater Reliability LLM Evaluation
Independent Researcher
MM-JudgeBias: A Benchmark for Evaluating Compositional Biases in MLLM-as-a-Judge
Apr 20, 2026Sua Lee, Sanghee Park, Jinbae ImLLM-as-a-Judge Multimodal Large Language Models
Seoul National University · NAVER Cloud AI · KAIST AI
Bias in the Loop: Auditing LLM-as-a-Judge for Software Engineering
Apr 18, 2026Zixiao Zhao, Amirreza Esmaeili, Fatemeh FardSoftware Engineering LLM-as-a-Judge
University of British Columbia, Canada
Context Over Content: Exposing Evaluation Faking in Automated Judges
Apr 16, 2026Manan Gupta, Inderjeet Nair, Lu Wang +1LLM-as-a-Judge Language Model Safety Evaluation
1BITS Pilani, India · University of Michigan
MLLM-as-a-Judge Exhibits Model Preference Bias
Apr 13, 2026Shuitsu Koyama, Yuiga Wada, Daichi Yashima +1Bias in LLM-as-a-Judge Multimodal Model Evaluation
Keio University, Japan
Cards Against LLMs: Benchmarking Humor Alignment in Large Language Models
Apr 9, 2026Yousra Fettach, Guillaume Bied, Hannu Toivonen +1LLM Alignment LLM Evaluation
Ghent University, Belgium · University of Helsinki, Finland
A Statistical Framework for Auditing Behavioral Dependence and Induced Bias in LLM Judges
Apr 8, 2026Chenchen Kuai, Jiwan Jiang, Zihao Zhu +8Correlated Errors in Language Models LLM Auditing
Texas A&M University · Marquette University · University of Utah
ReflectRM: Boosting Generative Reward Models via Self-Reflection within a Unified Judgment Framework
Apr 8, 2026Kai Qin, Liangxin Liu, Yu Liang +7Preference Modeling Self-Reflection in Language Model Agents
Shenzhen International Graduate School, Tsinghua University · Baidu Inc., Beijing, China
Self-Preference Bias in Rubric-Based Evaluation of Large Language Models
Apr 8, 2026José Pombal, Ricardo Rei, André F. T. MartinsBias in LLM-as-a-Judge LLM Evaluation
Sword Health · Instituto de Telecomunicações · Instituto Superior Técnico, Universidade de Lisboa +2
How Long Reasoning Chains Influence LLMs' Judgment of Answer Factuality
Apr 8, 2026Minzhu Tu, Shiyu Ni, Keping BiLLM-as-a-Judge LLM Evaluation
State Key Laboratory of AI Safety · Institute of Computing Technology, Chinese Academy of Sciences · Beijing University of Post and Telecommunications +1
Label Effects: Shared Heuristic Reliance in Trust Assessment by Humans and LLM-as-a-Judge
Apr 7, 2026Xin Sun, Di Wu, Sijing Qin +3Bias in LLM-as-a-Judge LLM-as-a-Judge
National Institute of Informatics (NII), Japan · University of Amsterdam, the Netherlands · Hitotsubashi University, Japan +3
A Scoping Review of LLM-as-a-Judge in Healthcare and the MedJUDGE Framework
Apr 3, 2026Chenyu Li, Zohaib Akhtar, Mingu Kwak +13LLM-as-a-Judge Clinical Language Model Evaluation
Department of Biomedical Informatics, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA · Department of Health Information Management, School of Health and Rehabilitation Sciences, University of Pittsburgh, Pittsburgh, PA, USA · OpenCura, Health Innovation Consortium +6

Jul 2, 2026J/K move · Enter open · S save

Ask the Right Comparison:Bias-Aware Bayesian Active Top- $k$ Ranking with LLM Judges

Jian Xu, Delu Zeng, John Paisley, Qibin Zhao

RIKEN iTHEMS · RIKEN AIP · South China University of Technology+1

Pairwise Comparison LLM-as-a-Judge Learning to Rank LLM Reranking Pairwise Ranking Language Model Bias Evaluation+2

Open paper AskPDF arXiv

Large language models (LLMs) are increasingly used as cheap, scalable judges that compare candidate outputs pairwise -- to rank responses, select models, or triage papers. Yet LLM judges are both noisy and systematically biased: they favor verbose or well-formatted answers and exhibit position effects, so simply aggregating their votes recovers a ranking of presentation, not of true quality. We study the practical goal of identifying the \topk{} items under a fixed comparison budget, and make two contributions. First, we cast judging as Bayesian inference over latent quality with explicit, judge-specific bias covariates (verbosity, position), regularized by a shrinkage prior so that the data decide which biases a given judge actually exhibits. Second, we introduce a \topk-aware active acquisition rule that chooses the next comparison to maximally reduce uncertainty about \topk{} membership, rather than about the full ranking. On a controlled benchmark with known ground-truth quality, judged by sixteen real LLMs spanning open and proprietary families (Llama, Qwen, Phi-4, GPT-4o-mini/5.1/5.5, Gemini, DeepSeek, and Claude Haiku/Sonnet/Opus), naive aggregation plateaus at a wrong \topk{} on biased judges regardless of budget, while our bias-aware model recovers it; \topk-aware acquisition reaches this ceiling with far fewer comparisons than round-robin or a global-uncertainty (D-optimal) rule. Bias is real but heterogeneous and capability-dependent: cheap and mid-tier judges carry a strong verbosity bias that our model corrects (lifting recall from

\sim$$0.5

0.6

0.84

1.0

), whereas the frontier judges we tested show little bias and already rank accurately, so bias-aware modeling changes little there.