Data to 14 Oct 2026

Evaluating Scoring Bias in LLM-as-a-Judge

arXiv 2506.22316 ↗First posted 27 Jun 2025cs.CLcited by 43

Not on the map. Why?

Abstract (arXiv)

The "LLM-as-a-Judge" paradigm, using Large Language Models (LLMs) as automated evaluators, is pivotal to LLM development, offering scalable feedback for complex tasks. However, the reliability of these judges is compromised by various biases. Existing research has heavily concentrated on biases in comparative evaluations. In contrast, scoring-based evaluations-which assign an absolute score and are often more practical in industrial applications-remain under-investigated. To address this gap, we undertake the first dedicated examination of scoring bias in LLM judges. We shift the focus from biases tied to the evaluation targets to those originating from the scoring prompt itself. We formally define scoring bias and identify three novel, previously unstudied types: rubric order bias, score ID bias, and reference answer score bias. We propose a comprehensive framework to quantify these biases, featuring a suite of multi-faceted metrics and an automatic data synthesis pipeline to create a tailored evaluation corpus. Our experiments empirically demonstrate that even the most advanced LLMs suffer from these substantial scoring biases. Our analysis yields actionable insights for designing more robust scoring prompts and mitigating these newly identified biases.

Key facts. Cited by 43 papers in the map (data through 14 Oct 2026). Builds on 38 of its 56 references that are in the map. First posted on arXiv 27 Jun 2025.

Builds on

38 of 56 references.

References · 38

2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
2025Qwen3 Technical Reportcited by 8,820
2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,050
2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
2024A Survey on LLM-as-a-Judgecited by 1,423
2023WizardLM: Empowering large pre-trained language models to follow complex instructionscited by 1,125
2023Large Language Models are not Fair Evaluatorscited by 1,085
2023RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedbackcited by 625
2023Prometheus: Inducing Fine-grained Evaluation Capability in Language Modelscited by 575
2024From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judgecited by 494
2024Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Modelscited by 446

26 more

2024Justice or Prejudice? Quantifying Biases in LLM-as-a-Judgecited by 368
2023Is ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agentscited by 360
2023PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimizationcited by 316
2023Large Language Models are Effective Text Rankers with Pairwise Ranking Promptingcited by 307
2023Evaluating Large Language Models at Evaluating Instruction Followingcited by 302
2024Self-Preference Bias in LLM-as-a-Judgecited by 258
2024Judging the Judges: Evaluating Alignment and Vulnerabilities in LLMs-as-Judgescited by 205
2022Leveraging Large Language Models for Multiple Choice Question Answeringcited by 203
2023FLASK: Fine-grained Language Model Evaluation based on Alignment Skill Setscited by 201
2023Benchmarking Foundation Models with Language-Model-as-an-Examinercited by 172
2024Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluatorscited by 149
2023Human-like Summarization Evaluation with ChatGPTcited by 126
2023PRD: Peer Rank and Discussion Improve Large Language Model based Evaluationscited by 114
2024A Comprehensive Survey on Process-Oriented Automatic Text Summarization with Exploration of LLM-Based Methodscited by 102
2023LLM-Eval: Unified Multi-Dimensional Automatic Evaluation for Open-Domain Conversations with Large Language Modelscited by 101
2024OffsetBias: Leveraging Debiased Data for Tuning Evaluatorscited by 94
2024The BiGGen Bench: A Principled Benchmark for Fine-grained Evaluation of Language Models with Language Modelscited by 93
2024Systematic Evaluation of LLM-as-a-Judge in LLM Alignment Tasks: Explainable Metrics and Diverse Prompt Templatescited by 72
2023CritiqueLLM: Towards an Informative Critique Generation Model for Evaluation of Large Language Model Generationcited by 66
2023T-SciQ: Teaching Multimodal Chain-of-Thought Reasoning via Mixed Large Language Model Signals for Science Question Answeringcited by 60
2023Evaluation Metrics in the Era of GPT-4: Reliably Evaluating Large Language Models on Sequence to Sequence Taskscited by 53
2024DALK: Dynamic Co-Augmentation of LLMs and KG to answer Alzheimer's Disease Questions with Scientific Literaturecited by 50
2024Instance-adaptive Zero-shot Chain-of-Thought Promptingcited by 26
2025Pairwise or Pointwise? Evaluating Feedback Protocols for Bias in LLM-Based Evaluationcited by 24
2025TRACT: Regression-Aware Fine-tuning Meets Chain-of-Thought Reasoning for LLM-as-a-Judgecited by 17
2023StrategyLLM: Large Language Models as Strategy Generators, Executors, Optimizers, and Evaluators for Problem Solvingcited by 13

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 6; 2026: 37.

43 papers.

Sort:

By area:

2025MCP-Bench: Benchmarking Tool-Using LLM Agents with Complex Real-World Tasks via MCP Serverscited by 109
2026Grading Scale Impact on LLM-as-a-Judge: Human-LLM Alignment Is Highest on 0-5 Grading Scalecited by 20
2026On Robustness and Chain-of-Thought Consistency of RL-Finetuned VLMscited by 11
2026SciVisAgentBench: A Benchmark for Evaluating Scientific Data Analysis and Visualization Agentscited by 10
2025ACE-RL: Adaptive Constraint-Enhanced Reward for Long-form Generation Reinforcement Learningcited by 9
2026OpenNovelty: An LLM-powered Agentic System for Verifiable Scholarly Novelty Assessmentcited by 8
2026Beyond Accuracy: Policy Invariance as a Reliability Test for LLM Safety Judgescited by 8
2025AWPO: Enhancing Tool-Use of Large Language Models through Adaptive Integration of Reasoning Rewardscited by 7
2026Am I More Pointwise or Pairwise? Revealing Position Bias in Rubric-Based LLM-as-a-Judgecited by 6
2026Are LLM Evaluators Really Narcissists? Sanity Checking Self-Preference Evaluationscited by 5
2026MM-JudgeBias: A Benchmark for Evaluating Compositional Biases in MLLM-as-a-Judgecited by 4
2026From Holistic Evaluation to Structured Criteria: Rubrics Across the Evolving LLM Landscapecited by 4

31 more

2025Pessimistic Verification for Open Ended Math Questionscited by 3
2026Beyond Tokens: Semantic-Aware Speculative Decoding for Efficient Inference by Probing Internal Statescited by 3
2026Label Effects: Shared Heuristic Reliance in Trust Assessment by Humans and LLM-as-a-Judgecited by 3
2026Auditing Multimodal LLM Raters: Central Tendency Bias in Clinical Ordinal Scoringcited by 3
2026Judging LLM-as-a-Judge: Concerning Rubric Artifacts in LLM-based Automated Text Generation Evaluationcited by 3
2025Readability Reconsidered: A Cross-Dataset Analysis of Reference-Free Metricscited by 2
2026Designing Reliable LLM-Assisted Rubric Scoring for Constructed Responses: Evidence from Physics Examscited by 2
2026More Choices, Fewer Decisions: Ordinal-Scale Bias in JEV-like Direct-Decision Modelscited by 2
2026CryptoAnalystBench: Failures in Multi-Tool Long-Form LLM Analysiscited by 1
2026How Uncertain Is the Grade? A Benchmark of Uncertainty Metrics for LLM-Based Automatic Assessmentcited by 1
2026Towards Provably Unbiased LLM Judges via Bias-Bounded Evaluationcited by 1
2026Implicit Grading Bias in Large Language Models: How Writing Style Affects Automated Assessment Across Math, Programming, and Essay Taskscited by 1
2026MemoNoveltyAgent: A Historical Research Memory-Aware Agent Workflow for Paper Novelty Assessmentcited by 1
2026SEAR: Schema-Based Evaluation and Routing for LLM Gatewayscited by 1
2026ODRPO: Ordinal Decompositions of Discrete Rewards for Robust Policy Optimizationcited by 1
2026PaperJury: Due-Process Review for Bounded LaTeX Revisioncited by 1
2026Are LLMs Positionally Consistent Ordinal Classifiers? A Systematic Evaluationcited by 1
2026Harness the Memory: A Holistic Evaluation of Memory Substrates in Memory Agentscited by 1
2026Rating the Raters: Rasch Measurement Theory for LLM Evaluationcited by 1
2025Knowledge-Graph Based RAG System Evaluation Frameworkcited by 0
2026A Systematic Approach for Large Language Models Debuggingcited by 0
2026MADRAG: Multi-Agent Debate with Retrieval-Augmented Generation for Training-Free Analytic Essay Scoringcited by 0
2026SPLIT: Cross-Lingual Empathy and Cultural Grounding in English and Ukrainian LLM Responsescited by 0
2026Automatic Hard Example Synthesis with Multi-Level Agentic Data Curationcited by 0
2026Mitigating Scoring Bias in LLM-as-a-Judge via Random Number Generationcited by 0
2026Mitigating Rubric Interference in LLM Judges via On-Policy Self-Distillationcited by 0
2026When Context Misleads: Intent-Guided Decoding for Robust Retrieval-Augmented Generationcited by 0
2026When Trust Meets Truth: Trust-Truth Separability in LLM-as-Judgecited by 0
2026MAWILE: Multi-Axis Workbench for Inspecting LLM Evaluatorscited by 0
2026RecToolBench: Benchmarking Recommendation-Specific Tool Orchestration under Fuzzy User Intentcited by 0
2026How Much Do LLM-as-a-Judge Design Choices Matter? A Systematic Comparison of Prompt Designs, Rating Scales, and Modelscited by 0

Similar, not cited

Close in wording, no citation either way