# Mitigating the Bias of Large Language Model Evaluation

arXiv: https://arxiv.org/abs/2409.16788 · HTML: https://mlsanity.com/paper/2409.16788/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 22 references that are in the map.
- First posted on arXiv 25 Sep 2024.

## Builds on (most cited first, 14 of 14)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Sparks of Artificial General Intelligence: Early experiments with GPT-4](https://mlsanity.com/paper/2303.12712/) (2023)
- [Calibrate Before Use: Improving Few-Shot Performance of Language Models](https://mlsanity.com/paper/2102.09690/) (2021)
- [A Survey on Evaluation of Large Language Models](https://mlsanity.com/paper/2307.03109/) (2023)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity](https://mlsanity.com/paper/2302.04023/) (2023)
- [AGIEval: A Human-Centric Benchmark for Evaluating Foundation Models](https://mlsanity.com/paper/2304.06364/) (2023)
- [Is ChatGPT a General-Purpose Natural Language Processing Task Solver?](https://mlsanity.com/paper/2302.06476/) (2023)
- [How Close is ChatGPT to Human Experts? Comparison Corpus, Evaluation, and Detection](https://mlsanity.com/paper/2301.07597/) (2023)
- [JudgeLM: Fine-tuned Large Language Models are Scalable Judges](https://mlsanity.com/paper/2310.17631/) (2023)
- [One Embedder, Any Task: Instruction-Finetuned Text Embeddings](https://mlsanity.com/paper/2212.09741/) (2022)
- [PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimization](https://mlsanity.com/paper/2306.05087/) (2023)
- [Evaluating Large Language Models at Evaluating Instruction Following](https://mlsanity.com/paper/2310.07641/) (2023)

## Built on by (most cited first, 17 of 17)

- [From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judge](https://mlsanity.com/paper/2411.16594/) (2024)
- [LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods](https://mlsanity.com/paper/2412.05579/) (2024)
- [CalibraEval: Calibrating Prediction Distribution to Mitigate Selection Bias in LLMs-as-Judges](https://mlsanity.com/paper/2410.15393/) (2024)
- [Calibrating LLM Judges: Linear Probes for Fast and Reliable Uncertainty Estimation](https://mlsanity.com/paper/2512.22245/) (2025)
- [Toward Robust LLM-Based Judges: Taxonomic Bias Evaluation and Debiasing Optimization](https://mlsanity.com/paper/2603.08091/) (2026)
- [Evaluating and Mitigating LLM-as-a-judge Bias in Communication Systems](https://mlsanity.com/paper/2510.12462/) (2025)
- [FairJudge: An Adaptive, Debiased, and Consistent LLM-as-a-Judge](https://mlsanity.com/paper/2602.06625/) (2026)
- [LLM Judges Can Be Too Generous When There Is No Reference Answer](https://mlsanity.com/paper/2607.12885/) (2026)
- [The Judge Who Never Admits: Hidden Shortcuts in LLM-based Evaluation](https://mlsanity.com/paper/2602.07996/) (2026)
- [Subliminal Signals in Preference Labels](https://mlsanity.com/paper/2603.01204/) (2026)
- [RM-Distiller: Exploiting Generative LLM for Reward Model Distillation](https://mlsanity.com/paper/2601.14032/) (2026)
- [Style Wins, Substance Loses: A Diagnosis of LLM-as-Judge in Idea Generation](https://mlsanity.com/paper/2608.01666/) (2026)
- [Query Symbolically or Retrieve Semantically? A Dataset and Method for Semi-Structured Question Answering](https://mlsanity.com/paper/2605.27164/) (2026)
- [Making Bias Non-Predictive: Training Robust LLM Reasoning via Reinforcement Learning](https://mlsanity.com/paper/2602.01528/) (2026)
- [Beyond the Resum'e: A Rubric-Aware Automatic Interview System for Information Elicitation](https://mlsanity.com/paper/2603.01775/) (2026)
- [Blinded Radiologist and LLM-Based Evaluation of LLM-Generated Japanese Translations of Chest CT Reports: Comparative Study](https://mlsanity.com/paper/2604.02207/) (2026)
- [DoGBench: Can Agents Meet Expert Standards for User-Facing Documentation?](https://mlsanity.com/paper/2609.39909/) (2026)

## Abstract (arXiv)

Recently, there has been a trend of evaluating the Large Language Model (LLM) quality in the flavor of LLM-as-a-Judge, namely leveraging another LLM to evaluate the current output quality. However, existing judges are proven to be biased, namely they would favor answers which present better superficial quality (such as verbosity, fluency) while ignoring the instruction following ability. In this work, we propose systematic research about the bias of LLM-as-a-Judge. Specifically, for closed-source judge models, we apply calibration to mitigate the significance of superficial quality, both on probability level and prompt level. For open-source judge models, we propose to mitigate the bias by contrastive training, with curated negative samples that deviate from instruction but present better superficial quality. We apply our methods on the bias evaluation benchmark, and experiment results show our methods mitigate the bias by a large margin while maintaining a satisfactory evaluation accuracy.