用辅助模型消除大模型评分中的偏见，让评估更公正。
Assistant-Guided Mitigation of Teacher Preference Bias in LLM-as-a-Judge
- 引入非偏向的助手模型补充训练数据，缓解教师偏好偏差
- 三阶段框架有效降低评分偏见，六项基准测试表现稳定
- 适合需要公平评估的AI生成内容研究者使用
LLM-as-a-Judge 利用 GPT-4 等大语言模型评估 LLM 生成回复质量，因其成本低且与人类评价高度一致而广受欢迎。然而，使用强大教师模型生成的评价数据训练代理判官模型时，会引入一个此前未被重视的问题：教师偏好偏差，即代理模型习得对教师模型回复的偏爱。为解决此问题，我们提出新范式，引入一个不偏向教师模型回复的额外助手模型，以补充训练数据。基于该设定，我们设计 AGDe-Judge，一种三阶段框架，用于从标签和反馈中同时去除偏差。大量实验表明，AGDe-Judge 能有效减轻教师偏好偏差，同时在六个评估基准上保持优异性能。代码已公开于 https://github.com/Liuz233/AGDe-Judge。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。