LLMs cannot reliably recognize their parametric knowledge boundaries and often hallucinate answers to outside-of-boundary questions. In this paper, we introduce MASH (Modeling Abstention via Selective Help-seeking), a training framework that readily extracts abstentions from LLMs. Our key idea is that any external help-seeking by an LLM, i.e. search tool use, can serve as a proxy for abstention if the external help (search) is appropriately penalized while also rewarding answer accuracy. MASH operationalizes this idea using reinforcement learning with a pay-per-search reward. We run experiments on three knowledge-intensive QA datasets. Our results show that MASH substantially improves upon the selective help-seeking performance of prior efficient search approaches; on multi-hop datasets, it improves answer accuracy by 7.6%. Furthermore, MASH demonstrates strong off-the-shelf abstention performance, showcasing behavior competitive with prior abstention methods that additionally require predetermining model knowledge boundaries to construct training data. Overall, we show MASH training effectively aligns search tool use with parametric knowledge, which can be successfully leveraged for making abstention decisions and efficient search tool use

翻译：大语言模型（LLM）无法可靠地识别其参数化知识边界，常对边界外问题产生幻觉回答。本文提出MASH（通过选择性求助建模弃权行为），一个可便捷提取LLM弃权输出的训练框架。核心思想在于：若在奖励回答准确性的同时，对LLM的外部求助行为（即检索工具使用）施加适当惩罚，则任何外部求助均可被视为弃权的代理信号。MASH通过引入按次付费搜索的强化学习机制实现该理念。我们在三个知识密集型问答数据集上开展实验。结果表明，MASH相较于先前高效搜索方法，在选择性求助性能上取得显著提升；在多跳数据集上，回答准确性提高7.6%。此外，MASH展现出优秀的即用型弃权性能，其表现与需要预判模型知识边界以构建训练数据的先验弃权方法相当。总体而言，我们证明MASH训练可有效将检索工具使用与参数化知识对齐，从而成功用于弃权决策与高效检索工具使用。