# Do LLM Evaluators Prefer Themselves for a Reason?

arXiv: https://arxiv.org/abs/2504.03846 · HTML: https://mlsanity.com/paper/2504.03846/

## Key facts

- Cited by 38 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 64 of its 71 references that are in the map.
- First posted on arXiv 4 Apr 2025.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)
- LLM alignment / RLHF / preference optimisation / [Human values and AI feedback alignment](https://mlsanity.com/alignment/human-values-and-ai-feedback-alignment/)

## Builds on (most cited first, 30 of 64)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [Solving Quantitative Reasoning Problems with Language Models](https://mlsanity.com/paper/2206.14858/) (2022)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [Large Language Monkeys: Scaling Inference Compute with Repeated Sampling](https://mlsanity.com/paper/2407.21787/) (2024)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback](https://mlsanity.com/paper/2309.00267/) (2023)
- [From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipeline](https://mlsanity.com/paper/2406.11939/) (2024)
- [Large Language Models Are Not Robust Multiple Choice Selectors](https://mlsanity.com/paper/2309.03882/) (2023)
- [Generative Verifiers: Reward Modeling as Next-Token Prediction](https://mlsanity.com/paper/2408.15240/) (2024)
- [Scalable agent alignment via reward modeling: a research direction](https://mlsanity.com/paper/1811.07871/) (2018)
- [Self-critiquing models for assisting human evaluators](https://mlsanity.com/paper/2206.05802/) (2022)

## Built on by (most cited first, 30 of 38)

- [One Token to Fool LLM-as-a-Judge](https://mlsanity.com/paper/2507.08794/) (2025)
- [Beyond the Surface: Measuring Self-Preference in LLM Judgments](https://mlsanity.com/paper/2506.02592/) (2025)
- [When Identity Skews Debate: Anonymization for Bias-Reduced Multi-Agent Reasoning](https://mlsanity.com/paper/2510.07517/) (2025)
- [Variation in Verification: Understanding Verification Dynamics in Large Language Models](https://mlsanity.com/paper/2509.17995/) (2025)
- [Let's Think in Two Steps: Mitigating Agreement Bias in MLLMs with Self-Grounded Verification](https://mlsanity.com/paper/2507.11662/) (2025)
- [Self-Preference Bias in Rubric-Based Evaluation of Large Language Models](https://mlsanity.com/paper/2604.06996/) (2026)
- [Quantifying and Mitigating Self-Preference Bias of LLM Judges](https://mlsanity.com/paper/2604.22891/) (2026)
- [WaymoQA: A Multi-View Visual Question Answering Dataset for Safety-Critical Reasoning in Autonomous Driving](https://mlsanity.com/paper/2511.20022/) (2025)
- [Mitigating Self-Preference by Authorship Obfuscation](https://mlsanity.com/paper/2512.05379/) (2025)
- [Am I More Pointwise or Pairwise? Revealing Position Bias in Rubric-Based LLM-as-a-Judge](https://mlsanity.com/paper/2602.02219/) (2026)
- [MAS-ProVe: Understanding the Process Verification of Multi-Agent Systems](https://mlsanity.com/paper/2602.03053/) (2026)
- [Are LLM Evaluators Really Narcissists? Sanity Checking Self-Preference Evaluations](https://mlsanity.com/paper/2601.22548/) (2026)
- [The Hidden Signal of Verifier Strictness: Controlling and Improving Step-Wise Verification via Selective Latent Steering](https://mlsanity.com/paper/2605.20745/) (2026)
- [Aligning Large Language Models via Fully Self-Synthetic Data](https://mlsanity.com/paper/2510.06652/) (2025)
- [Assistant-Guided Mitigation of Teacher Preference Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2505.19176/) (2025)
- [Quantifying Label-Induced Bias in Large Language Model Self- and Cross-Evaluations](https://mlsanity.com/paper/2508.21164/) (2025)
- [Breaking the Mirror: Activation-Based Mitigation of Self-Preference in LLM Evaluators](https://mlsanity.com/paper/2509.03647/) (2025)
- [On the Shelf Life of Fine-Tuned LLM-Judges: Future-Proofing, Backward-Compatibility, and Question Generalization](https://mlsanity.com/paper/2509.23542/) (2025)
- [When LLMs Benchmark Themselves: Deconstructing Self-Bias in Automated Evaluation](https://mlsanity.com/paper/2509.26600/) (2025)
- [CodeSimpleQA: Scaling Factuality in Code Large Language Models](https://mlsanity.com/paper/2512.19424/) (2025)
- [Self-Attribution Bias: When AI Monitors Go Easy on Themselves](https://mlsanity.com/paper/2603.04582/) (2026)
- [Neural Models and Language Model Prompting for the Multidimensional Evaluation of Open-Ended Conversations](https://mlsanity.com/paper/2509.00841/) (2025)
- [Extreme Self-Preference in Language Models](https://mlsanity.com/paper/2509.26464/) (2025)
- [Social Caption: Evaluating Social Understanding in Multimodal Models](https://mlsanity.com/paper/2601.14569/) (2026)
- [Quantifying Self-Preservation Bias in Large Language Models](https://mlsanity.com/paper/2604.02174/) (2026)
- [CoEval: Ranking Language Models for Custom Tasks Without Labeled Data or Trustworthy Benchmarks](https://mlsanity.com/paper/2606.03650/) (2026)
- [Mitigating Judgment Preference Bias in Large Language Models through Group-Based Polling](https://mlsanity.com/paper/2510.08145/) (2025)
- [Understanding LLM Evaluator Behavior: A Structured Multi-Evaluator Framework for Merchant Risk Assessment](https://mlsanity.com/paper/2602.05110/) (2026)
- [MLLM-as-a-Judge Exhibits Model Preference Bias](https://mlsanity.com/paper/2604.11589/) (2026)
- [Self- and Other-Labels Induce Bidirectional Bias in LLM Judges](https://mlsanity.com/paper/2608.18091/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are increasingly used as automatic evaluators in applications such as benchmarking, reward modeling, and self-refinement. Prior work highlights a potential self-preference bias where LLMs favor their own generated responses, a tendency often intensifying with model size and capability. This raises a critical question: Is self-preference harmful, or does it simply reflect the genuinely higher-quality outputs of stronger models? Answering this has been difficult as prior works mostly relied on subjective tasks that lack an objective ground truth, meaning that either preference can be reasonably justified. To address this ambiguity, we investigate self-preference using verifiable benchmarks (mathematical reasoning, factual knowledge, code generation) that allow objective ground-truth assessment. This enables us to distinguish harmful (favoring objectively worse responses) from legitimate (favoring genuinely superior ones) self-preference. Our large-scale experiments across 7 model families reveal three key findings: (1) While stronger models exhibit greater self-preference, much of this preference aligns with objectively superior performance, indicating stronger models prefer themselves mostly legitimately. (2) Harmful self-preference persists when evaluator models err as generators, and stronger models display more pronounced harmful self-preference bias when they do err. This suggests stronger models struggle more to recognize when they are wrong. (3) Inference-time scaling strategies, such as generating a long Chain-of-Thought before evaluation, effectively reduce the harmful self-preference. Additionally, we experiment with LMArena and show that our findings extend beyond verifiable benchmarks to real-world, subjective domains. These results provide a more nuanced understanding of LLM-based evaluation and practical insights for improving its reliability.