解释LLM-as-a-Judge策略：通过可验证的全局解释
Interpreting LLM-as-a-Judge Policies via Verifiable Global Explanations
AI总结：
本文提出从LLM-as-a-Judge中提取可验证全局策略的方法，包含CLoVE和GloVE两种算法，在七个基准数据集上验证了高忠实度、鲁棒性，并通过用户研究评估了理解与满意度。
摘要：
使用LLM评估文本，即LLM-as-a-judge，正越来越多地被大规模用于增强甚至取代人工标注。因此，我们必须理解这样做可能存在的偏见和风险。在这项工作中，我们提出了一种从LLM-as-a-Judge中提取基于高级概念全局策略的方法。我们的方法包含两种算法：1）CLoVE（对比局部可验证解释），它生成可验证的、基于概念的、对比性的局部解释；2）GloVE（全局可验证解释），它使用迭代聚类、总结和验证将局部规则浓缩为全局策略。我们在七个用于内容危害检测的标准基准数据集上评估了GloVE。我们发现提取的全局策略对LLM-as-a-Judge的决策高度忠实。此外，我们还评估了全局策略对文本扰动和对抗性攻击的鲁棒性。最后，我们进行了一项用户研究，以评估用户对全局策略的理解和满意度。
英文摘要：
Using LLMs to evaluate text, that is, LLM-as-a-judge, is increasingly being used at scale to augment or even replace human annotations. As such, it is imperative that we understand the potential biases and risks of doing so. In this work, we propose an approach for extracting high-level concept-based global policies from LLM-as-a-Judge. Our approach consists of two algorithms: 1) CLoVE (Contrastive Local Verifiable Explanations), which generates verifiable, concept-based, contrastive local explanations and 2) GloVE (Global Verifiable Explanations), which uses iterative clustering, summarization and verification to condense local rules into a global policy. We evaluate GloVE on seven standard benchmarking datasets for content harm detection. We find that the extracted global policies are highly faithful to decisions of the LLM-as-a-Judge. Additionally, we evaluated the robustness of global policies to text perturbations and adversarial attacks. Finally, we conducted a user study to evaluate user understanding and satisfaction with global policies.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
