Large Language Models (LLMs) can infer sensitive attributes such as gender or age from indirect cues like names and pronouns, potentially biasing recommendations. While several debiasing methods exist, they require access to the LLMs' weights, are computationally costly, and cannot be used by lay users. To address this gap, we investigate implicit biases in LLM Recommenders (LLMRecs) and explore whether prompt-based strategies can serve as a lightweight and easy-to-use debiasing approach. We contribute three bias-aware prompting strategies for LLMRecs. To our knowledge, this is the first study on prompt-based debiasing approaches in LLMRecs that focuses on group fairness for users. Our experiments with 3 LLMs, 4 prompt templates, 9 sensitive attribute values, and 2 datasets show that our proposed debiasing approach, which instructs an LLM to be fair, can improve fairness by up to 74% while retaining comparable effectiveness, but might overpromote specific demographic groups in some cases.

翻译：大型语言模型（LLM）能够从姓名、代词等间接线索推断性别、年龄等敏感属性，可能导致推荐结果产生偏见。尽管已有多种去偏方法，但它们通常需要访问LLM的权重参数，计算成本高昂，且普通用户难以使用。为填补这一空白，本研究探究了LLM推荐系统（LLMRecs）中存在的隐性偏见，并探索提示策略能否作为一种轻量级、易使用的去偏方法。我们提出了三种面向LLMRecs的偏见感知提示策略。据我们所知，这是首个针对LLMRecs中用户群体公平性的提示式去偏方法研究。通过使用3种LLM、4种提示模板、9类敏感属性值及2个数据集进行实验，结果表明：指导LLM保持公平性的去偏方法最高可提升74%的公平性指标，同时保持相当的推荐效能，但在某些情况下可能导致特定人口群体的过度推荐。