# Don't Judge Code by Its Cover: Exploring Biases in LLM Judges for Code Evaluation

arXiv: https://arxiv.org/abs/2505.16222 · HTML: https://mlsanity.com/paper/2505.16222/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 45 references that are in the map.
- First posted on arXiv 22 May 2025.

## Builds on (most cited first, 28 of 28)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2004.04906/) (2020)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [A Survey on LLM-as-a-Judge](https://mlsanity.com/paper/2411.15594/) (2024)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [CodeBLEU: a Method for Automatic Evaluation of Code Synthesis](https://mlsanity.com/paper/2009.10297/) (2020)
- [CodeT: Code Generation with Generated Tests](https://mlsanity.com/paper/2207.10397/) (2022)
- [MLLM-as-a-Judge: Assessing Multimodal LLM-as-a-Judge with Vision-Language Benchmark](https://mlsanity.com/paper/2402.04788/) (2024)
- [JudgeLM: Fine-tuned Large Language Models are Scalable Judges](https://mlsanity.com/paper/2310.17631/) (2023)
- [JudgeBench: A Benchmark for Evaluating LLM-based Judges](https://mlsanity.com/paper/2410.12784/) (2024)
- [LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks](https://mlsanity.com/paper/2406.18403/) (2024)
- [CodeNet: A Large-Scale AI for Code Dataset for Learning a Diversity of Coding Tasks](https://mlsanity.com/paper/2105.12655/) (2021)
- [SPoC: Search-based Pseudocode to Code](https://mlsanity.com/paper/1906.04908/) (2019)
- [An Empirical Study of the Non-determinism of ChatGPT in Code Generation](https://mlsanity.com/paper/2308.02828/) (2023)
- [CodeBERTScore: Evaluating Code Generation with Pretrained Models of Code](https://mlsanity.com/paper/2302.05527/) (2023)
- [Can LLMs Replace Human Evaluators? An Empirical Study of LLM-as-a-Judge in Software Engineering](https://mlsanity.com/paper/2502.06193/) (2025)
- [Out of the BLEU: how should we assess quality of the Code Generation models?](https://mlsanity.com/paper/2208.03133/) (2022)
- [ICE-Score: Instructing Large Language Models to Evaluate Code](https://mlsanity.com/paper/2304.14317/) (2023)
- [Does BLEU Score Work for Code Migration?](https://mlsanity.com/paper/1906.04903/) (2019)
- [CodeJudge-Eval: Can Large Language Models be Good Judges in Code Understanding?](https://mlsanity.com/paper/2408.10718/) (2024)
- [Are LLM-Judges Robust to Expressions of Uncertainty? Investigating the effect of Epistemic Markers on LLM-based Evaluation](https://mlsanity.com/paper/2410.20774/) (2024)
- [Large Language Models as Test Case Generators: Performance Evaluation and Enhancement](https://mlsanity.com/paper/2404.13340/) (2024)
- [Benchmarking Adversarial Robustness to Bias Elicitation in Large Language Models: Scalable Automated Assessment with LLM-as-a-Judge](https://mlsanity.com/paper/2504.07887/) (2025)
- [Benchmarking LLMs' Judgments with No Gold Standard](https://mlsanity.com/paper/2411.07127/) (2024)
- [LLMs can be easily Confused by Instructional Distractions](https://mlsanity.com/paper/2502.04362/) (2025)
- [CodeSift: An LLM-Based Reference-Less Framework for Automatic Code Validation](https://mlsanity.com/paper/2408.15630/) (2024)
- [A Systematic Literature Review on the Impact of Formatting Elements on Code Legibility](https://mlsanity.com/paper/2208.12141/) (2022)

## Built on by (most cited first, 18 of 18)

- [CodeJudgeBench: Benchmarking LLM-as-a-Judge for Coding Tasks](https://mlsanity.com/paper/2507.10535/) (2025)
- [From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligence](https://mlsanity.com/paper/2511.18538/) (2025)
- [Belief in Authority: Impact of Authority in Multi-Agent Evaluation Framework](https://mlsanity.com/paper/2601.04790/) (2026)
- [ISO-Bench: Can Coding Agents Optimize Real-World Inference Workloads?](https://mlsanity.com/paper/2602.19594/) (2026)
- [Uncovering Systematic Failures of LLMs in Verifying Code Against Natural Language Specifications](https://mlsanity.com/paper/2508.12358/) (2025)
- [On LLMs' Internal Representation of Code Correctness](https://mlsanity.com/paper/2512.07404/) (2025)
- [Aletheia: What Makes RLVR For Code Verifiers Tick?](https://mlsanity.com/paper/2601.12186/) (2026)
- [Bias in the Loop: Auditing LLM-as-a-Judge for Software Engineering](https://mlsanity.com/paper/2604.16790/) (2026)
- [LLM Judges Can Be Too Generous When There Is No Reference Answer](https://mlsanity.com/paper/2607.12885/) (2026)
- [Measuring and Exploiting Contextual Bias in LLM-Assisted Security Code Review](https://mlsanity.com/paper/2603.18740/) (2026)
- [Scaffold, Not Vocabulary? A Controlled, Two-Tier, Pre-Registered Study of a Popperian Code-Generation Skill](https://mlsanity.com/paper/2606.06454/) (2026)
- [Probing Multimodal Large Language Models on Cognitive Biases in Chinese Short-Video Misinformation](https://mlsanity.com/paper/2601.06600/) (2026)
- [OptiHive: Ensemble Selection for LLM-Based Optimization via Statistical Modeling](https://mlsanity.com/paper/2508.02503/) (2025)
- [Judging What We Cannot Solve: A Consequence-Based Approach for Oracle-Free Evaluation of Research-Level Math](https://mlsanity.com/paper/2602.06291/) (2026)
- [Themis: Training Robust Multilingual Code Reward Models for Flexible Multi-Criteria Scoring](https://mlsanity.com/paper/2605.00754/) (2026)
- [SkillEval: Decomposing Agent Skill Quality into Interpretable Signals](https://mlsanity.com/paper/2608.06891/) (2026)
- [CriterAlign: Criterion-Centric Rationale Alignment for Code Preference Judging](https://mlsanity.com/paper/2605.19665/) (2026)
- [Style, Not Self: Surface Cues Explain Zero-Shot Code Attribution by Large Language Models](https://mlsanity.com/paper/2609.30048/) (2026)

## Abstract (arXiv)

With the growing use of large language models(LLMs) as evaluators, their application has expanded to code evaluation tasks, where they assess the correctness of generated code without relying on reference implementations. While this offers scalability and flexibility, it also raises a critical, unresolved question: Can LLM judges fairly and robustly evaluate semantically equivalent code with superficial variations? Functionally correct code often exhibits variations-such as differences in variable names, comments, or formatting-that should not influence its correctness. Yet, whether LLM judges can reliably handle these variations remains unclear. We present the first comprehensive study of this issue, defining six types of potential bias in code evaluation and revealing their systematic impact on LLM judges. Across five programming languages and multiple LLMs, we empirically demonstrate that all tested LLM judges are susceptible to both positive and negative biases, resulting in inflated or unfairly low scores. Moreover, we observe that LLM judges remain vulnerable to these biases even when prompted to generate test cases before scoring, highlighting the need for more robust code evaluation methods.