mlsanityData to 14 Oct 2026

Assistant-Guided Mitigation of Teacher Preference Bias in LLM-as-a-Judge

arXiv 2505.19176 ↗First posted 25 May 2025cs.CLcited by 3

On the map: LLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment

Abstract (arXiv)

LLM-as-a-Judge employs large language models (LLMs), such as GPT-4, to evaluate the quality of LLM-generated responses, gaining popularity for its cost-effectiveness and strong alignment with human evaluations. However, training proxy judge models using evaluation data generated by powerful teacher models introduces a critical yet previously overlooked issue: teacher preference bias, where the proxy judge model learns a biased preference for responses from the teacher model. To tackle this problem, we propose a novel setting that incorporates an additional assistant model, which is not biased toward the teacher model's responses, to complement the training data. Building on this setup, we introduce AGDe-Judge, a three-stage framework designed to debias from both the labels and feedbacks in the training data. Extensive experiments demonstrate that AGDe-Judge effectively reduces teacher preference bias while maintaining strong performance across six evaluation benchmarks. Code is available at https://github.com/Liuz233/AGDe-Judge.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 21 of its 29 references that are in the map. First posted on arXiv 25 May 2025.

Builds on

21 of 29 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025Do LLM Evaluators Prefer Themselves for a Reason?4 Apr
  2. 2024LLM Evaluators Recognize and Favor Their Own Generations15 Apr
  3. 2024Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference7 Mar
  4. 2024Debating with More Persuasive LLMs Leads to More Truthful Answers9 Feb
  5. 2023Prometheus: Inducing Fine-grained Evaluation Capability in Language Models12 Oct

Other references · 16

  1. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  2. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  3. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
  4. 2024Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluatorscited by 897
  5. 2024Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Modelscited by 446
  6. 2024Justice or Prejudice? Quantifying Biases in LLM-as-a-Judgecited by 368
  7. 2023JudgeLM: Fine-tuned Large Language Models are Scalable Judgescited by 367
  8. 2024Humans or LLMs as the Judge? A Study on Judgement Biasescited by 323
  9. 2023PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimizationcited by 316
  10. 2024Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judgecited by 269
  11. 2023Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewardscited by 243
  12. 2024LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Taskscited by 243
4 more
  1. 2024Meta-Rewarding Language Models: Self-Improving Alignment with LLM-as-a-Meta-Judgecited by 204
  2. 2023Benchmarking Foundation Models with Language-Model-as-an-Examinercited by 172
  3. 2023PRD: Peer Rank and Discussion Improve Large Language Model based Evaluationscited by 114
  4. 2024Beyond Scalar Reward Model: Learning Generative Judge from Preference Datacited by 32

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025Mitigating Judgment Preference Bias in Large Language Models through Group-Based Pollingcited by 1
  2. 2026Project Synapse: A Hierarchical Multi-Agent Framework with Hybrid Memory for Autonomous Resolution of Last-Mile Delivery Disruptionscited by 1
  3. 2026AGC-Bench: Measuring Artificial General Creativitycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Beyond the Surface: Measuring Self-Preference in LLM Judgments
  2. 2026Quantifying and Mitigating Self-Preference Bias of LLM Judges
  3. 2026Toward Robust LLM-Based Judges: Taxonomic Bias Evaluation and Debiasing Optimization
  4. 2024Self-Preference Bias in LLM-as-a-Judge
  5. 2026Mitigating LLM biases toward spurious social contexts using direct preference optimization
  6. 2024Direct Judgement Preference Optimization
  7. 2025Evaluating Scoring Bias in LLM-as-a-Judge
  8. 2026FairJudge: An Adaptive, Debiased, and Consistent LLM-as-a-Judge